CA2167091A1

CA2167091A1 - B7-2: ctl a4/cd 28 counter receptor

Info

Publication number: CA2167091A1
Application number: CA002167091A
Authority: CA
Inventors: Gordon J. Freeman; Lee M. Nadler; Gary S. Gray; Edward Greenfield
Original assignee: Individual
Current assignee: Dana Farber Cancer Institute Inc; Genetics Institute LLC
Priority date: 1993-07-26
Filing date: 1994-07-26
Publication date: 1995-02-02
Also published as: AU9699198A; WO1995003408A1; JPH09500788A; EP0711345A1; JP2009060905A; AU7405294A

Abstract

Nucleic acids encoding novel CTLA4/CD28 ligands which costimulate T cell activation are disclosed. In one embodiment, the nucleic acid has a sequence which encodes a B lymphocyte antigen, B7-2. Preferably , the nucleic acid is a DNA molecule comprising at least a portion of a nucleotide sequence shown in Figure 8, SEQ ID NO: 1 or Figure 14, SEQ ID NO 23.
The nucleic acid sequences of the invention can be integrated into various expression vectors, which in turn direct the synthesis of the corresponding proteins or peptides in a variety of hosts, particularly eukaryotic cells, such as mammalian and insect cell culture. Also disclosed are host cells transformed to produce proteins or peptides encoded by the nucleic acid sequences of the invention and isolated proteins and peptides which comprise at least a portion of a novel B lymphocyte antigen. Proteins and peptides described herein can be administered to subjects to enhance or suppress T cell-mediated immune responses.

Description

WO 95/03~8 ~ 1 6 ~ ~ 9 ~ PCT/US94/08423 B 7 - 2: CTL A4/CD 28 COUNTER RECEPTOR

Govern~ent Fnndi~
Work described herein was supported under CA-40216-08 awarded by the National 5 Institutes of Health. The U.S. government therefore may have certain rights in this invention.

Rarl~round of the Inv~ntion To induce antigen-specific T cell activation and clonal expansion, two signals provided by antigen-presentin~ cells (APCs) must be delivered to the surface of resting T
lymphocytes (Jenkins, M. and Schwartz, R. (1987)J. Exp. Med. 165, 302-319; Mueller, D.L., et al. (1990) J. Immunol. 144, 3701-3709; Williams, I.R. and Unanue, E.R. (1990) J.
Immunol. 145, 85-93). The first signal, which confers specificity to the imm~1ne response, is mediated ~ia the T cell receptor (TCR) following recognition of foreign antigenic peptide presented in the context of the major histocompatibility complex (MHC). The second signal, termed costim~ tion, induces T cells to proliferate and become functional (Schwartz, R.H.
(1990) Science 2~, 1349-1356). Costimulation is neither antigen-specific, nor MHC
restricted and is thought to be provided by one or more distinct cell surface molecules expressed by APCs (Jenkins, M.K., et al. (1988) J. Immunol. 140, 3324-3330; Linsley, P.S., etal. (l991)J. ~p. Med. 173, 721-730; Gimmi, C.D., etal., (1991)Proc. Natl. Acad. Sci.
USA. 88, 6575-6579; Young, J.W., et al. (1992) J. Clin. Invest. 90, 229-237; Koulova, L., et al. (1991) J. Exp. Med. 173, 759-762; Reiser, H., et al. (1992) Proc. Natl. Acad. Sci. USA. 89, 271-275; van-Seventer, G.A., et al. (1990) J. Immunol. 144, 4579-4586; LaSalle, J.M., et al., (1991) J. Immunol. 147, 774-80; Dustin, M.I., et al., (1989) J. E~cp. Med. 169, 503; Armitage, R.J., et al. (1992) Nature ~1, 80-82; Liu, Y., et al. (1992) J. E~p. Med. ~ 75, 437-445).
Considerable evidence suggests that the B7 protein, expressed on APCs, is one such critical costimulatory molecule (Linsley, P.S., et al., (1991) J. Exp. Med. 173, 721-730;
Gimmi, C.D., et al., (1991) Proc. Natl. Acad. Sci USA. $8, 6575-6579; Koulova, L., et al., (1991) J. E~cp. Med. L73, 759-762; Reiser, H., et al. (1992) Proc. Natl. Acad. Sci. USA. 89, 271 -275; Linsley, P.S. et al. (1990) Proc. Natl. Acad. Sci. USA. 87, 5031 -5035; Freeman, G.J.
et al. (1991) J. ~p. Med. 174,625-631.). B7 is the counter-receptor for two ligands expressed on T lymphocytes. The first ligand, terrned CD28, is constitutively expressed on resting T cells and increases after activation. After ~ign~linp through the T cell receptor, ligation of CD28 induces T cells to proliferate and secrete IL-2 (Linsley, P.S., et al. (1991) J
E~p. Med. 173, 721-730; Girnmi, C.D., et al. (1991) Proc. Nafl. ,4cad. Sci. USA. 88, 6575-6579; Thompson, C.B., et al. (1989) Proc. Natl. Acad. Sci. USA. 86, 1333-1337; June, C.H., et al. (1990) Immunol. Today. ~1, 211-6; Harding, F.A., et al. (1992) Nature. 356, 607-609.).
The second ligand, termed CTLA4 is homologous to CD28 but is not expressed on resting T
cells and appears following T cell activation (Brunet, J.F., et al., (1987) Nature ~, 267-270). DNA sequences encoding the human and murine CTLA4 protein are described in WO 9~/03408 PCT/US94/08423 ~67~91 Dariavich, et al. (1988) Eur. J. Immunol. 18(12), 1901-1905; Brunet, J.F., et al. (1987) supra;
Brunet, J.F. et al. (1988) Immunol. Rev. 103:21-36; and Freeman, G.J.? et al. (1992) J.
Immunol. 149, 3795-3801. Although B7 has a higher affinity for CTLA4 than for CD28 (Linsley, P.S., et al., (1991) J. Exp. Med 174, 561-569), the function of CTL~;4 is still 5 unknown.
The importance of the B7:CD28/CTLA4 costimulatory pathway has been demonstrated in vitro and in several in vivo model systems. Blockade of this costimulatory pathway results in the development of antigen specific tolerance in murine and hl-m~n~
systems (Harding, F.A., et al. (1992) Nature. ~, 607-609; Lenschow, D.J., et al. (1992) Science. ~, 789-792, Turka, L.A., et al. (1992) Proc. Natl. Acad. Sci. USA. 89, 11102-11105; Gimmi, C.D., et al. (1993) Proc. Natl. Acad. Sci USA ~, 6586-6590; Boussiotis, V., et al. (1993) J: Exp. Med. 178, 1753-1763). Conversely, expression of B7 by B7 negative murine tumor cells induces T-cell mediated specific immunity accompanied by tumor rejection and long lasting protection to tumor challenge (Chen, L., et al. (1992) Cell 71, 1093-1102; Townsend, S.E. and Allison, J.P. (1993) Science ~,~, 368-370; Baskar, S., et al.
(1993) Proc. Natl. Acad. Sci. 9Q, 5687-5690.). Therefore, manipulation of the B7:CD28/CTLA4 pathway offers great potential to stim~ te or suppress immune responses in hllm~n~

20 ~I-mm~ry of the Inver tion This invention pertains to isolated nucleic acids encoding novel molecules whichcostimlll~te T cell activation. Preferred cosfim~ tQry molecules include antigens on the surface of B lymphocytes, professional antigen ples~ in~ cells (e.g., monocytes, dendritic cells, Langerhan cells) and other cells (e.g., keratinocytes, endothelial cells, astrocytes, 25 fibroblasts, oligodendrocytes) which present antigen to immune cells, and which bind either CTLA4, CD28, both CTLA4 and CD28 or other known or as yet undefined receptors onimmune cells. Such costim~ tQry molecules are referred to herein as CTLA4/CD28 binding counter-receptors or B lymphocyte antigens, and are capable of providing costimulation to activated T cells to thereby induce T cell proliferation and/or cytokine secretion. Preferred B
30 lymphocyte antigens include B7-2 and B7-3 and soluble fragments or derivatives thereof which bind CTLA4 and/or CD28 and have the ability to inhibit or induce costimulation of immllne cells. In one embodiment, an isolated nucleic acid which encodes a peptide having the activity of the hDan B7-2 B Iymphocyte antigen is provided. Preferably, the nucleic acid is a cDNA molecule having a nucleotide sequence encoding human B7-2, as shown in 35 Figure 8 (SEQ ID NO: 1). In another embodiment, the nucleic acid is a cDNA molecule having a nucleotide sequence encoding murine B7-2, as shown in Figure 14 (SEQ IDNO:22).

wo 95,03408 ~ ~ ~ 7 0 ~ :L PCT/US94/08423 The invention also features nucleic acids which encode a peptide having B7-2 activity and at least about 50%, more preferably at least about 60% and most preferably at least about 70% homologous with an amino acid sequence shown in Figure 8 (SEQ ID NO:2) or anamino acid sequence shown in Figure 14 (SEQ ID NO:23). Nucleic acids which encode peptides having B7-2 activity and at least about 80%, more preferably at least about 90%, more preferably at least about 95% and most preferably at least about 98% or at least about 99% homologous with an amino acid sequence shown in Figure 8 (SEQ ID NO:2) or anamino acid sequence shown in Figure 14 (SEQ ID NO:23) are also within the scope of the invention. In another embodiment, the peptide having B7-2 activity is encoded by a nucleic acid which hybridizes under high or low stringency conditions to a nucleic acid which encodes a peptide having an amino acid sequence of Figure 8 (SEQ ID NO:2) or a peptide having an amino acid sequence shown in Figure 14 (SEQ ID NO:23).
The invention further pertains to an isolated nucleic acid comprising a nucleotide sequence encoding a peptide having B7-2 activity and having a length of at least 20 amino acid residues. Peptides having B7-2 activity and con~i~tin~ of at least 40 amino acid residues in length, at least 60 amino acid residues in length, at least 80 amino acid residues in length, at least 100 amino acid residues in length or at least 200 or more amino acid residues in length are also ~,vithin the scope of this invention. Particularly preferred nucleic acids encode a peptide having B7-2 activity, a length of at least 20 amino acid residues or more and at least 50% or greater homology (preferably at least 70%) with a sequence shown in Figure 8 (SEQ
ID NO:2).
In one preferred embodiment, the invention features an isolated DNA encoding a peptide having B7-2 activity and an amino acid sequence represented by a formula:

Xn~Y~Zm In the formula, Y consists e~enti~lly of amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2). Xn and Zm are additional amino acid residue(s) linked to Y by an amide bond. Xn and Zm are arnino acid residues selected from amino acid residuescontiguous to Y in the amino acid sequence shown in Figure 8 (SEQ ID NO:2). Xn is amino acid residue(s) selected from amino acids contiguous to the amino terminus of Y in the " sequence shown in Figure 8 (SEQ ID NO:2), i.e., selected from arnino acid residue 23 to 1.
Zm is amino acid residue(s) selected from amino acids contiguous to the carboxy terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2), i.e., selected from amino acid residue 246 to 329. According to the formula, n is a number from 0 to 23 (n=0-23) and m is a number from 0 to 84 (m=0-84). A particularly preferred DNA encodes a peptide having an 9 ~ ~

amino acid sequence represented by the formula Xn-Y-Zm, where Y is amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2) and n=0 and m=0.
The invention also features an isolated DNA encoding a B7-2 fusion protein whichincludes a nucleotide sequence encoding a first peptide having B7-2 activity and a nucleotide 5 sequence encoding a second peptide corresponding to a moiety that alters the solubility, binding affinity, stability or valency of the f1rst peptide. Preferably, the first peptide having B7-2 activity includes an extracellular domain portion of the B7-2 protein (e.g., about amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2)) and the second peptide is an immunoglobulin constant region, for example, a human Cyl or C~4 domain, 10 including the hinge, CH2 and CH3 region, to produce a B7-2 imml-noglobulin fusion protein (B7-2Ig)(see Capon et al. (1989) Nature ~1, 525-531 and Capon U.S. 5,116,964).
The nucleic acids obtained in accordance with the present invention can be inserted into various expression vectors, which in turn direct the synthesis of the corresponding protein or peptides in a variety of hosts, particularly eucaryotic cells, such as m~mm~ n and 15 insect cell culture, and procaryotic cells such as E. coli. Expression vectors within the scope of the invention comprise a nucleic acid encoding at least one peptide having the activity of a novel B Iymphocyte antigen as described herein, and a promoter operably linked to the nucleic acid sequence. In one embodiment, the t;~ s~ion vector contains a DNA encoding a peptide having the activity of the B7-2 antigen and a DNA encoding a peptide having the 20 activity of another B Iymphocyte antigen, such as the previously characterized B7 activation antigen, referred to herein as B7-1. Such t;~ ;s~ion vectors can be used to transfect host cells to thereby produce proteins and peptides, including fusion proteins, encoded by nucleic acids as described herein.
Nucleic acid probes useful for assaying a biological sample for the presence of B cells 25 e~les~ing the B Iymphocyte antigens B7-2 and B7-3 are also within the scope of the invention.
The invention further pertains to isolated peptides having the activity of a novel B
Iymphocyte antigen, including the B7-2 and B7-3 protein antigens. A preferred peptide having B7-2 activity is produced by recombinant expression and comprises an amino acid 30 sequence shown in Figure 8 (SEQ ID NO: 2). Another preferred peptide having B7-2 activity comprises an amino acid sequence shown in Figure 14 (SEQ ID NO:23). A particularly preferred peptide having the activity of the B7-2 antigen includes at least a portion of the mature form of the protein, such as an extracellular domain portion (e.g., about amino acid residues 24-245 of SEQ ID NO:2) which can be used to enhance or suppress T-cell mediated 35 immune responses in a subject. Other preferred peptides having B7-2 activity include peptides having an amino acid sequence represented by a formula:
.

wo gs/03408 2 ~ 6 7 ~ 9 1 PCT/US94/08423 Xn~Y~Zm In the formula, Y is amino acid residues selected from the group consisting of: amino acid residues 55-68 ofthe sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 81-89 ofthe sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 128-142 ofthe sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 160-169 ofthe sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 188-200 of the sequence shown in Figure 8 (SEQ ID NO:2); and amino acid residues 269-282 of the sequence shown in Figure 8 (SEQ ID NO:2). In the formula Xn and Zm are additional amino acid residue(s) lin~ed to 10 Y by an amide bond and are selected from amino acid residues contiguous to Y in the amino acid sequence shown in Figure 8 (SEQ ID NO:2). Xn is amino acid residue(s) selected from amino acids contiguous to the amino terminl-~ of Y in the sequence shown in Figure 8 (SEQ
ID NO:2). Zm is arnino acid residue(s) selected from amino acids contiguous to the carboxy terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2). According to the formula, 15 n is a number from 0 to 30 (n=0-30) and m is a number from 0 to 30 (m=0-30).
Fusion proteins or hybrid fusion proteins including a peptide having the activity of a novel B lymphocyte antigen (e.g., B7-2, B7-3) are also featured. For example, a fusion protein comprising a first peptide which includes an extracellular domain portion of a novel B
lymphocyte antigen fused to second peptide, such as an immunoglobulin constant region, that alters the solubility, binding affinity, stability and/or valency of the first peptide are provided.
In one embodiment, a fusion protein is produced comprising a first peptide which includes amino acid residues of an extracellular domain portion of the B7-2 protein joined to a second pepide which includes amino acid residues of a sequence corresponding to the hinge, CH2 and CH3 regions of C~1 or C~4 to form a B7-2Ig fusion protein. In another embodiment, a hybrid fusion protein is produced compri~in~ a first peptide which includes an extracellular domain portion of the B7-1 antigen and an extracellular domain portion of the B7-2 antigen and a second peptide which includes amino acid residues corresponding to the hinge, CH2 and CH3 of C~1 (see e.g., Linsley et al. (1991) ~ E xp. Med. 1783 :721 -730; Capon et al.
(1989) Nature ~, 525-531, and Capon U.S. 5,116,964).
Isolated peptides and fusion proteins of the invention can be a-lmini~tered to a subject to either upregulate or inhibit the expression of one or more B Iymphocyte antigens or block the ligation of one or more B Iymphocyte antigens to their natural ligand on immune cells, such as T cells, to thereby provide enhancement or su~up~es~ion of cell-mediated immune responses in vivo.
Another embodiment of the invention provides antibodies, preferably monoclonal antibodies, specifically reactive with a peptide of a novel B lymphocyte antigen or fusion protein as described herein. Preferred antibodies are anti-human B7-2 monoclonal antibodies ' WO 95/03408 7 ~ 6- PCT/US94/08423 produced by hybridoma cells HF2.3D1, HA5.2B7 and HA3.1F9. These hybridoma cells have been deposited with the American Type Culture Collection at ATCC Accession No._ (EIF2.3D1), ATCC Accession No. (HA5.2B7)7 and ATCC Accession No.
(HA3.1F9).
A still further aspect of the invention involves the use of the nucleic acids of the r invention, especially the cDNAs, to enhance the immlln~genicity of a m~mm~ n cell. In preferred embodiments, the m~mm~ n cell is a tumor cell, such as a sarcoma, a lymphoma, a melanoma, a neuroblastoma, a leukemia or a carcinoma, or an antigen presenting cell, such as a macrophage, which is transfected to allow expression of a peptide having the activity of a novel B lymphocyte antigen of the invention on the surface of the cell. Macrophages that express a peptide having the activity of a B lymphocyte antigen, such as the B7-2 antigen, can be used as antigen presentin~ cells, which, when pulsed with an ~ iate pathogen-related antigen or tumor antigen, enhance T cell activation and immllne stimulation.
~mm~ n cells can be transfected with a suitable expression vector cont~inin~ a nucleic acid encoding a peptide having the activity of a novel B Iymphocyte antigen, such as the B7-2 antigen, ex vivo and then introduced into the host m~mm~l, or alternatively, cells can be transfected with the gene in vivo via gene therapy techniques. For example, the nucleic acid encoding a peptide having B7-2 activity can be transfected alone, or in combination with nucleic acids encoding other costimlll~tQry molecules. In enhancing the immlmogenicity of tumors which do not express Class I or Class II MHC molecules, it may be beneficial to additionally transfect a~prop,iate class I or II genes into the ms~mm~ n cells to be transfected with a nucleic acid encoding a peptide having the activity of a B lymphocyte antigen, as described herein.
The invention also provides methods for inducing both general immunosuppression and antigen-specific tolerance in a subject by,-for example, blocking the functional interaction ofthe novel B lymphocyte antigens ofthe invention, e.g., B7-2 and B7-3, to their natural ligand(s) on T cells or other immllne system cells, to thereby block co-stim~ tion through the receptor-ligand pair. In one embodiment, inhibitory molecules that can be used to block the interaction of the natural human B7-2 antigen to its natural ligands (e.g., CTLA4 and CD28) include a soluble peptide having B7-2 binding activity but lacking the ability to costimlll~te immllne cells, antibodies that block the binding of B7-2 to its ligands and fail to deliver a co-stimnlsltQry signal (so called "blocking antibodies", such as blocking anti-B7-2 antibodies), B7-2-Ig fusion proteins, which can be produced in accordance with the te~c.hing.
of the present invention, as well as soluble forms of B7-2 receptors~ such as CTLA4Ig or CD28Ig. Such blocking agents can be used alone or in combination with agents which block interaction of other costimul~tory molecules with their natural ligands (e.g., anti-B7 antibody). Inhibition of T cell responses and induction of T cell tolerance according to the

2 16 7 0 91 PCT/US94/08423 ..

methods described herein may be useful prophylactically, in preventing transplantation rejection (solid organ, skin and bone marrow) and graft versus host disease, especially in allogeneic bone marrow transplantation. The methods of the invention may also be useful fherapeutically, in the treatment of autoimmllne ~ e~es, allergy and allergic reactions, transplantation rejection, and established graft versus host disease in a subject.
Another aspect of the invention features methods for upregulating immune responses by delivery of a costim~ tory signal to T cells through use of a stimulatory form of B7-2 antigen, which include soluble, multivalent forms of B7-2 protein, such as a peptide having B7-2 activity and B7-2 fusion proteins. Delivery of a stimul~tory form of B7-2 in conjunction with antigen may be useful prophylactically to enhance the efficacy of vaccination against a variety of pathogens and may also be useful therapeutically to upregulate an immune response against a particular pathogen during an infection or against a tumor in a tumor-bearing host.
The invention also features methods of identifying molecules which can inhibit either the interaction of B Iymphocyte antigens, e.g., B7-2, B7-3, with their receptors or ~ r~le with intracellular si~n~llin~ through their receptors. Methods for identifying molecules which can modulate the expression of B lymphocyte antigens on cells are also provided. In addition, methods for identifying cytokines produced in response to costimlll~tion of T cells by novel B Iymphocyte antigens are within the scope of the invention.
Brief Description of the nr.~
Figure lA-B are graphic le~l`eSr~ 1 ions of the responses of CD28+ T cells, as ~e~ce~l by 3H-thymidine incorporation or IL-2 secretion, to costim~ tion provided by either B7 (B7-1) transfected CHO cells (panel a) or syngeneic activated B Iymphocytes (panel b) cultured in media, anti-CD3 alone, or anti-CD3 in the presence of the following monoclonal antibodies or recombinant proteins: aB7 (133, anti-B7-1); CTLA4Ig; Fab a CD28; control Ig fusion protein (isotype control for CTLA4Ig); or aB5 (anti-B5, the isotype control for anti-B7-1).
Figure 2A-C are graphs of log fluorescence intensity of cell surface expression of B7-1 on splenic B cells activated with surface immunoglobulin (sIg) crocslinkin~;. The total (panel a), B7-1 positive (B7-1+, panel b) and B7-1 negative (B7-1-, panel c) activated B cells " were stained with anti-B7-1 monoclonal antibody (133) and fluoroscein isothiocyanate (FITC) labeled goat anti-mouse immunoglobulin and analyzed by flow cytometry.
Figure 3A-B are graphic representations of the responses of CD28+ T cells, as 35 assessed by 3H-thymidine incorporation and IL-2 secretion, to costimulation provided by B7-1+ (panel a) or B7-1- (panel b) activated syngeneic B Iymphocytes cultured in media, anti-CD3 alone, or anti-CD3 in the presence of the following monoclonal antibodies or WO 9s/03408 ~16 ~ ~ 91 PCT/USg4/08423 recombinant proteins: aBB-I (133, anti-B7-1 and anti-B7-3); aB7 (anti-B7-1); CTLA4Ig;
Fab aCD28, control Ig fusion protein or aB5 (anti-B5).
Figure 4 is a graphic lel)lest:lltalion of the cell surface expression of B7-1, B7-3 and total CTLA4 counter-receptors on fractionated B7-1+ and B7-1- activated B Iymphocytes.
- 5 Figure S is a graphic representation of temporal surface expression of B7-1 (CTLA4Ig and mAbs BB-l and 133), B7-3 (CTLA4Ig and mAb BBl) and B7-2 (CTLA4Ig) counter-receptors on splenic B cells activated by sIg cro~linking.
Figure 6is a graphic representation of temporal surface expression of B7- 1 - (CTLA4Ig and mAbs BB-l and 133), B7-3 (CTLA4Ig and mAb BBl) and B7-2 (CTLA4Ig) counter-receptors on splenic B cells activated by MHC class II cros~linkinf~.
Figure 7A-B are graphic representations of the response of CD28+ T cells, as assessed by 3H-thymidine incorporation and IL-2 secretion, to costimulation provided by syngeneic B
lymphocytes activated by sIg cro~.~linking for 24 hours (panel a) or 48 hours (panel b) and cultured in media, anti-CD3 alone, or anti-CD3 in the presence of the following monoclonal antibodies or recombinant protein: aB7(133, anti-B7-1); aBBl (anti-B7-1, anti-B7-3) CTLA4Ig; Fab aCD28; and aB5(anti-BS).
Figure 8 is the nucleotide and ~le~ cecl amino acid sequence of the human B
lymphocyte antigen B7-2 (hB7-2-clone29).
Figure 9 is a graphic representation of COS cells transfected with control plasmid (pCDNAI), plasmid expressing B7-1 (B7-1), or plasmid ~;x~ules~ g B7-2 (B7-2) stained with either control mAb (IgM), anti-B7-1 (mAbs 133 and BB-l), recombinant protein CTLA4Ig, or isotype matched control Ig protein followed by the applopliate second FITC labelled immllnoglobulin and analyzed by flow cytometry.
Figure lOA-B show RNA blot analyses of B7-2 ~ lession in unstimulated and anti-Ig activated human spenic B cells and cell lines (panel a) and human myelomas (panel b).
Figure 11 is a graphic representation of the proliferation of CD28+ T cells, as ~se~ecl by 3H-thymidine incorporation or IL-2 secretion, to submitogenic stiml-l~tion with phorbol myristic acid (PMA) and COS cells transfected with vector alone or vectors directing the expression of either B7- 1 or B7-2.
Figure 12 is a graphic representation of the inhibition by mAbs and recombinant proteins of the proliferation of CD28+ T cells, as assessed by 3H-thymidine incorporation and IL-2 secretion, to stimulation by PMA and COS cells transfected with vector alone (vector), or with a vector expressing B7-1 (B7-1) or B7-2 (B7-2). Inhibition studies were performed with the addition of either no antibody (no mAb), anti-B7 mAb 133 (133), anti-B7 mAb BB-1 (BB1), anti-BS mAb (BS), Fab fragment of anti-CD28 (CD28 Fab), CTLA4Ig (CTLA4Ig), or Ig control protein (control Ig) to the PMA stimulated COS cell admixed CD28+ T cells.

wo 95/0340~ 2 ~ 6 7 ~ ~1 PCT/US94/08423 Figure 13 shows the sequence homology between the human B7-2 protein (h B7-2) deduced amino acid sequence (SEQ ID NO: 2) and the amino acid sequence of both the human B7-1 protein (h B7-1) (SEQ ID NO: 28 and 29) and the murine B7-1 protein (m B7) (SEQ ID NO: 30 and 31).
~igure 14 is the nucleotide and deduced amino acid sequence of the murine B7-2 antigen (mB7-2) (SEQ ID NO: 22 and 23).
Figure 15 is a graphic representation of the competitive inhibition of binding of biotinylated-CTLA4Ig to immobilized B7-2 Ig by B7 family-Ig fusion proteins. The Ig fusion proteins e~mined as competitors were: full-length B7-2 (hB7.2), full-length B7-1 (hB7.1), the variable region-like domain of B7-2 (hB7.2V) or the constant region-like domain of B7-2 (hB7.2C).
Figure 16A-B are graphic representations of the competitive inhibition of binding of biotinylated-B7-1-Ig (panel A) or B7-2-Ig (panel B) to immobilized CTLA4-Ig by increasing concentrations of unlabelled B7-1-Ig (panel A) or B7-2-Ig (panel B). The experimentally determined ICso values are indicated in the upper right corner of the panels.
Figure 17 depicts flow cytometric profiles of cells stained with an anti-hB7-2 monoclonal antibody, HA3. lF9. Cells stained with the antibody were CHO cells transfected to express human B7-2 (CHO-hB7.2), NIH 3T3 cells transfected to express human B7-2 (3T3-hB7.2) and conkol transfected NIH 3T3 cells (3T3-neo). The anti-hB7.2 antibody B70 was used as a positive control.
Figure 18 depicts flow cytometric profiles of cells stained with an anti-hB7-2 monoclonal antibody, HA5.2B7. Cells stained with the antibody were CHO cells transfected to express human B7-2 (CHO-hB7.2), NIH 3T3 cells transfected to express human B7-2 (3T3-hB7.2) and control transfected NIH 3T3 cells (3T3-neo). The anti-hB7.2 antibody B70 was used as a positive control.
Figure 19 depicts flow cytometric profiles of cells stained with an anti-hB7-2 monoclonal antibody, H~2.3D1. Cells stained with the antibody were CHO cells transfected to express human B7-2 (CHO-hB7.2), NIH 3T3 cells transfected to express human B7-2 (3T3-hB7.2) and control transfected NIH 3T3 cells (3T3-neo). The anti-hB7.2 antibody B70 was used as a positive control.
Figure 20 is a graphic representation of tumor cell growth (as measured by tumorsize) in mice following transplantation of J558 plasmacytoma cells or J558 plasmacytoma cells transfected to express B7-1 (J558-B7.1) or B7-2 (JS58-B7.2).

I)etailed Description of the Inv*ntion In addition to the previously characterized B lymphocyte activation antigen B7 (referred to herein as B7-1), human B lymphocytes express other novel molecules which WO 95/03408 2 ~ ~ 7 ~ 9 ~ PCT/US94/08423 costim~ te T cell activation. These costimulatory molecules include antigens on the surface of B lymphocytes, professional antigen presenting cells (e.g., monocytes, dendritic cells, Langerhan cells) and other cells (e.g.7 keratinocytes, endothelial cells, astrocytes, fibroblasts, oligodendrocytes) which present antigen to imml-ne cells, and which bind either CTLA4, 5 CD28, both CTLA4 and CD28 or other known or as yet undefined receptors on immllne cells. Costim~ tory molecules within the scope of the invention are referred to herein as CTLA4/CD28 ligands (counter-receptors) or B lymphocyte antigens. Novel B lymphocyte antigens which provide cotimulation to activated T cells to thereby induce T cell proliferation and/or cytokine secretion include the B7-2 (human and murine) and the B7-3 antigens 10 described and characterized herein.
The B Iymphocyte antigen B7-2 is expressed by human B cells at about 24 hours following stimlll~tion with either anti-immunoglobulin or anti-MHC class II monoclonal antibody. The B7-2 antigen induces detectable IL-2 secretion and T cell proliferation. At about 48 to 72 hours post activation, human B cells express both B7-1 and a third CTLA4 15 counter-receptor, B7-3, identified by a monoclonal antibody BB-l, which also binds B7-1 (Yokochi, T., et al. (1982) J. Immunol. 128, 823-827). The B7-3 antigen is also expressed on B7-1 negative activated B cells and can costimlll~te T cell proliferation without detectable IL-2 production, indicating that the B7-1 and B7-3 molecules are distinct. B7-3 is expressed on a wide variety of cells including activated B cells, activated monocytes, dendritic cells, 20 Langerhan cells and keratinocytes. At 72 hours post B cell activation, the expression of B7-1 and B7-3 begins to decline. The presence of these costimlll~tory molecules on the surface of activated B lymphocytes indicates that T cell costimulation is regulated, in part, by the temporal expression of these molecules following B cell activation.
Accordingly, one aspect of this invention pertains to isolated nucleic acids comprising 25 a nucleotide sequence encoding a novel costiml~ Qry molecule, such as the B lymphocyte antigen, B7-2, fragments of such nucleic acids, or equivalents thereo The term "nucleic acid" as used herein is intended to include such fragments or equivalents. The term "equivalent" is int~ncle~l to include nucleotide sequences encoding functionally equivalent B
lymphocyte antigens or functionally equivalent peptides having an activity of a novel B
30 lymphocyte antigen, i.e., the ability to bind to the natural ligand(s) of the B lymphocyte antigen on immllne cells, such as CTLA4 and/or CD28 on T cells, and inhibit (e.g., block) or stimulate (e.g., enhance) immune cell costimulation. Such nucleic acids are considered equivalents ofthe human B7-2 nucleotide sequence provided in Figure 8 (SEQ ID NO:l) and the murine B7-2 nucleotide sequence provided in Figure 14 (SEQ ID NO:22) and are within ..
35 the scope of this invention.
In one embodiment, the nucleic acid is a cDNA encoding a peptide having an activity of the B7-2 B Iymphocyte antigen. Preferably, the nucleic acid is a cDNA molecule WO 95/03408 ~ 9 L PCT/US94/08423 consisting of at least a portion of a nucleotide sequence encoding human B7-2~ as shown in Figure 8 (SEQ ID NO: 1) or at least a portion of a nucleotide sequence encoding murine B7-2, as shown in Figure 14 (SEQ ID NO:22). A preferred portion of the cDNA molecule of Figure 8 (SEQ ID NO: 1) or Figure 14 (SEQ ID NO:22) includes the coding region of the molecule.
In another embodiment, the nucleic acid of the invention encodes a peptide having an activity of B7-2 and comprising an amino acid sequence shown in Figure 8 (SEQ ID NO:2) or Figure 14 (SEQ ID NO:23). Preferred nucleic acids encode a peptide having B7-2 activity and at least about 50% homology, more preferably at least about 60% homology and most preferably at least about 70% homology with an amino acid sequence shown in Figure 8 (SEQ ID NO:2). Nucleic acids which encode peptides having B7-2 activity and at least about 90%, more preferably at least about 95%, and most preferably at least about 98-99%
homologous with a sequence set forth in Figure 8 (SEQ ID NO:2) are also within the scope of the invention. Homology refers to sequence similarity between two peptides having the activity of a novel B lymphocyte antigen, such as B7-2, or between two nucleic acid molecules. Homology can be rlett~rmined by comp~ring a position in each sequence which may be aligned for purposes of comparison. When a position in the compared sequences is occupied by the same nucleotide base or amino acid, then the molecules are homologous at that position. A degree (or percentage) of homology between sequences is a function of the number of m~tc.hing or homologous positions shared by the sequences.
Another aspect of the invention provides a nucleic acid which hybridizes under high or low skingency conditions to a nucleic acid which encodes a peptide having all or a portion of an amino acid sequence shown in Figure 8 (SEQ ID NO:2) or a peptide having all or a portion of an amino acid sequence shown in Figure 14 (SEQ ID NO:23). Appropriatestringency conditions which promote DNA hybridization, for example, 6.0 x sodiumchloride/sodium cikate (SSC) at about 45C, followed by a wash of 2.0 x SSC at 50C are known to those skilled in the art or can be found in Current Protocols in Molecular Biology, John Wiley & Sons, N.Y. (1989), 6.3.1-6.3.6. For example, the salt concenkation in the wash step can be selected from a low stringency of about 2.0 x SSC at 50C to a high stringency of about 0.2 x SSC at 50C. In addition, the temperature in the wash step can be increased from low skingency conditions at room temperature, about 22C to high stringency conditions, at about 65C.
Isolated nucleic acids encoding a peptide having an activity of a novel B lymphocyte antigen, as described herein, and having a sequence which differs from nucleotide sequence shown in Figure 8 (SEQ ID NO:l) or Figure 14 (SEQ ID NO:22) due to degeneracy in the genetic code are also within the scope of the invention. Such nucleic acids encode functionally equivalent peptides (e.g., a peptide having B7-2 activity) but differ in sequence ~16709`1 ~

from the sequence of Figure 8 or Figure 14 due to degeneracy in the genetic code. For example, a number of amino acids are clecign~ted by more than one triplet. Codons that specify the same amino acid, or synonyms (for example, CAU and CAC are synonyms for histidine) may occur due to degeneracy in the genetic code. As one example, bNA sequence 5 polymorphisms within the nucleotide sequence of a B7-2 (especially those within the third base of a codon) may result in "silent" mutations in the DNA which do not affect the amino acid encoded. However, it is expected that DNA sequence polymorphisms that do lead to changes in the amino acid sequences of the B7-2 antigen will exist within a population. It will be appreciated by one skilled in the art that these variations in one or more nucleotides 10 (up to about 3-4% of the nucleotides) of the nucleic acids encoding peptides having the activity of a novel B lymphocyte antigen may exist among individuals within a population due to natural allelic variation. Any and all such nucleotide variations and resnltin~ amino acid polymorphisms are within the scope of the invention. Furthermore, there may be one or more isoforms or related, cross-reacting family members of the novel B Iymphocyte antigens 15 described herein. Such isoforms or family members are defined as proteins related in - function and amino acid sequence to a B lymphocyte antigen (e.g., the B7-2 antigen), but encoded by genes at dirrelenl loci.
A "fragment" of a nucleic acid encoding a novel B lymphocyte antigen is defined as a nucleotide sequence having fewer nucleotides than the nucleotide sequence encoding the 20 entire amino acid sequence of the B lymphocyte antigen and which encodes a peptide having an activity of the B Iymphocyte antigen (i.e., the ability to bind to the natural ligand(s) of the B lymphocyte antigen on immllne cells, such as CTLA4 and/or CD28 on T cells and either stim--l~te or inhibit immllne cell costimnl~tion). Thus, a peptide having B7-2 activity binds CTLA4 and/or CD28 and stimulates or inhibits a T cell mediated immune response, as 25 evidenced by, for example, cytokine production and/or T cell proliferation by T cells that - have received a primary activation signal. In one embodiment, the nucleic acid fragment encodes a peptide of the B7-2 antigen which retains the ability of the antigen to bind CTLA4 and/or CD28 and deliver a costim~ tory signal to T Iymphocytes. In another embodiment, the nucleic acid fragment encodes a peptide including an extracellular portion of the human 30 B7-2 antigen (e.g., approximately amino acid residues 24-245 of the sequence provided in Figure 8 (SEQ ID NO:2)) which can be used to bind CTLA4 and/or CD28 and, in monovalent form, inhibit costimulation, or in multivalent form, induce or enhance costim~ tion.
Preferred nucleic acid fragments encode peptides of at least 20 amino acid residues in 35 length, preferably at least 40 amino acid residues and length, and more preferably at least 60 amino acid residues in length. Nucleic acid fragments which encode peptides of at least 80 amino acid residues in length, at least 100 amino acid residues in length, and at least 200 or WO 9~/03408 2 ~ 6 7 0 91 PCT/US94/08423 more amino acids in length are also within the scope of the invention. Particularly plef~ d nucleic acid fragments encode a peptide having the activity of human B7-2 and an amino acid sequence represented by a formula:

Xn-Y-Zm In the fomula, Y comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2). Xn and Zm are additional amino acid residue(s) linked to Y by an amide bond. Xn and Zm are selected from amino acid residues contiguous to Y in the amino acid sequence shown in Figure 8 (SEQ ID NO:2). In the formula, Xn is amino acid residue(s) selected from amino acids contiguous to the amino terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2), i.e., from amino acid residue 23 to 1. Zm is amino acid residue(s) selected from amino acids contiguous to the carboxy terminlls of Y in the sequence shown in Figure 8 (SEQ ID NO:2), i.e., from amino acid residue 246 to 329. In addition, in the formula, n is a number from 0 to 23 (n=0-23) and m is a number from 0 to 84 (m=0-84). A
particularly ~rer~lled peptide has an amino acid sequence represented by the formula Xn-Y-Zm as above, where n=0 and m=0.
Nucleic acid fragments within the scope of the invention include those capable of hybridizing with nucleic acid from other animal species for use in screening protocols to detect novel proteins that are cross-reactive with the B lymphocyte antigens described herein.
These and other fr~gment~ are described in detail herein. Generally, the nucleic acid encoding a fragment of a B lymphocyte antigen will be selected from the bases coding for the mature protein, however, in some instances it may be desirable to select all or part of a fragment or fragments from the leader sequence or non-coding portion of a nucleotide sequence. Nucleic acids within the scope of the invention may also contain linker sequences, modified restriction endonuclease sites and other sequences useful for molecular cloning, t;x~l~ssion or purification of recombinant protein or fragments thereof. These and other modifications of nucleic acid sequences are described in further detail herein.
A nucleic acid encoding a peptide having an activity of a novel B lymphocyte antigen, such as the B7-2 antigen, may be obtained from mRNA present in activated B lymphocytes.
It should also be possible to obtain nucleic acid sequences encoding B lymphocyte antigens from B cell genomic DNA. For example, the gene encoding the B7-2 antigen can be cloned from either a cDNA or a genomic library in accordance with protocols herein described. A
, cDNA encoding the B7-2 antigen can be obtained by isolating total mRNA from an 35 ~propl,ate cell line. Double stranded cDNAs can then prepared from the total mRNA.
Subsequently, the cDNAs can be inserted into a suitable plasmid or viral (e.g., bacteriophage) vector using any one of a number of known techniques. Genes encoding novel B lymphocyte WO 95/03408 ~ 7 ~ ~ ~ PCT/US94/08423 antigens can also be cloned using established polymerase chain reaction techniques in accordance with the nucleotide sequence information provided by the invention. The nucleic acids of the invention can be DNA or RNA. A p,~erell~ed nucleic acid is a cDNA encoding the human B7-2 antigen having the sequence depicted in Figure 8 (SEQ ID NO:1). Another 5 preferred nucleic acid is a cDNA encoding the murine B7-2 antigen having the sequence shown on Figure 14 (SEQ ID NO:22).
This invention further pertains to ~x~les~ion vectors cont~ininp a nucleic acid encoding at least one peptide having the activity of a novel B lymphocyte antigen, as described herein, operably linked to at least one regulatory sequence. "Operably linked" is 10 inten~le~l to mean that the nucleotide acid sequence is linked to a regulatory sequence in a manner which allows t;~,c;s~ion of the nucleotide sequence (e.g., in cis or trans). Regulatory sequences are art-recognized and are selected to direct expression of the desired protein in an appropriate host cell. Accordingly, the term regulatory sequence includes promoters, enhancers and other expression control elements. Such regulatory sequences are known to 15 those skilled in the art or one described in Goeddel, Gene Expression Technology: Methods in Enzymology 185, ~c~ tnic Press, San Diego, CA (1990). It should be understood that the design of the expression vector may depend on such factors as the choice of the host cell to be transfected and/or the type of protein desired to be expressed. In one embodiment, the .lession vector includes a nucleic acid encoding at least a portion of the B7-2 protein, such 20 as an extracellular domain portion. In another embodiment, the expression vector includes a DNA encoding a peptide having an activity of the B7-2 antigen and a DNA encoding a peptide having an activity of another B lymphocyte antigen, such as B7-1. cDNAs encoding the human B7-1 and mouse B7-1 antigens are shown in SEQ ID NO:28 and SEQ ID NO:30, respectively. The ~iedl~cecl amino acid sequences of these antigens are also shown in SEQ ID
25 NO:29 and SEQ ID NO:3 1, respectively. Such expression vectors can be used to transfect cells to thereby produce proteins or peptides, including fusion proteins or peptides encoded by nucleic acid sequences as described herein. These and other embodiments are described in further detail herein.
The invention also features methods of producing peptides having an activity of a 30 novel B lymphocyte antigen. For example, a host cell transfected with a nucleic acid vector directing ~ ,ession of a nucleotide sequence encoding a peptide having an activity of the B7-2 protein can be cultured in a medium under applopliate conditions to allow expression of the peptide to occur. In addition, one or more expression vectors cont~inin~ DNA encoding a - peptide having an activity of B7-2 and DNA encoding another peptide, such as a peptide 35 having an activity of a second B Iymphocyte antigen (e.g., B7-1, B7-3) can be used to transfect a host cell to coexpress these peptides or produce fusion proteins or peptides. In one embodiment, a recombinant ~ ssion vector cont~ininp DNA encoding a B7-2 fusion protein is produced. A B7-2 fusion protein can be produced by recombinant expression of a nucleotide sequence encoding a first peptide having B7-2 activity and a nucleotide sequence encoding second peptide corresponding to a moiety that alters the solubility, affinity, stability or valency of the first peptide, for example, an immunoglobulin constant region. Preferably, - 5 the first peptide consists of a portion of the extracellular domain of the human B7-2 antigen (e.g., approximately amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID
NO:2)). The second peptide can include an immnnoglobulin constant region, for example, a human C~1 domain or C~4 domain (e.g., the hinge, CH2 and CH3 regions of human Ig~
or human IgC~4, see e.g., Capon et al. US 5,116,964, incol~o~led herein by reference). A
reslllting B7-2Ig fusion protein may have altered B7-2 solubility, binding affinity, stability and/or valency (i.e., the number of binding sites available per molecule) and may increase the efficiency of protein purification. Fusion proteins and peptides produced by recombinant technique may be secreted and isolated from a mixture of cells and medium cont~ining the protein or peptide. Alternatively, the protein or peptide may be retained cytoplasmically and the cells harvested, lysed and the protein isolated. A cell culture typically includes host cells, media and other byproducts. Suitable mediums for cell culture are well known in the art.
Protein and peptides can be isolated from cell culture medium, host cells, or both using techniques known in the art for purifying proteins and peptides. Techniques for transfecting host cells and purifying proteins and peptides are described in further detail herein.
Particularly preferred human B7-2Ig fusion proteins include the extracellular domain portion or variable region-like domain of human B7-2 coupled to an immunoglobulin constant region. The immllnoglobulin constant region may contain genetic modifications which reduce or elimin~te effector activity inherent in the immlmnglobulin structure. For example, DNA encoding the extracellular portion of human B7-2 (hB7-2), as well as DNA
encoding the variable region-like domain of human B7-2 (hB7.2V) or the constant region-like domain of human B7-2 (hB7.2C) can be joined to DNA encoding the hinge, CH2 and CH3 regions of human IgC~1 and/or IgCy4 modified by site directed mutagenesis. The Lion and chara~;le. ;,~ ion of these fusion proteins is described in detail in Example 7.
Transfected cells which express peptides having an activity of one or more B
lymphocyte antigens (e.g., B7-2, B7-3) on the surface of the cell are also within the scope of this invention. In one embodiment, a host cell such as a COS cell is transfected with an ~x~lts~ion vector directing the ~ c;s~ion of a peptide having B7-2 activity on the surface of the cell. Such a transfected host cell can be used in methods of identifying molecules which inhibit binding of B7-2 to its counter-receptor on T cells or which interfere with intracellular ~i~n~ling of costim~ tion to T cells in response to B7-2 interaction. In another embodiment, a tumor cell such as a sarcoma, a melanoma, a lellkemi~ a lymphoma, a carcinoma or a neuroblastoma is transfected with an e~ ion vector directing the expression of at least one WO 95/03408 21~ PCT/US94/08423 peptide having the activity of a novel B lymphocyte antigen on the surface of the tumor cell.
In some instances, it may be beneficial to transfect a tumor cell to coexpress major histocompatibility complex (MHC) proteins, for example MHC class II a and ,B chain proteins or an MHC class I a chain protein, and, if necessary, a ~2 microglobulin protein.
5 Such transfected tumor cells can be used to induce tumor immllnity in a subject. These and other embo-liment~ are described in further detail herein.
The nucleic acid sequences of the invention can also be chemically synthesi7.?d using standard techniques. Various methods of chemically synthesizing polydeoxynucleotides are known, including solid-phase synthesis which, like peptide synthesis, has been fully 10 automated in commercially available DNA synthesi7~rs (See e.g., Itakura et al. U.S. Patent No. 4,598,049; Caruthers et al. U.S. Patent No. 4,458,066, and Itakura U.S. Patent Nos.
4,401,796 and 4,373,071, incorporated by reference herein).
Another aspect of the invention pertains to isolated peptides having an activity of a novel B lymphocyte antigen (e.g., B7-2, B7-3). A peptide having an activity of a B
15 Iymphocyte antigen may differ in amino acid sequence from the B Iymphocyte antigen, such as the human B7-2 sequence depicted in Figure 8 (SEQ ID NO:2), or murine B7-2 sequence depicted in Figure 14 (SEQ ID NO:22), but such differences result in a peptide which functions in the same or similar manner as the B Iymphocyte antigen or which has the same or similar characteristics of the B Iymphocyte antigen. For example, a peptide having an 20 activity of the B7-2 protein is defined herein as a peptide having the ability to bind to the natural ligand(s) of the B7-2 protein on immllne cells, such as CLTA4 and/or CD28 on T
cells and either stimulate or inhibit immune cell costimlll~tion. Thus, a peptide having B7-2 activity binds CTLA4 and/or CD28 and stimlll~tes or inhibits a T cell mediated immune response (as evidenced by, for example, cytokine production andlor proliferation by T cells 25 that have received a primary activation signal). One embodiment provides a peptide having B7-2 binding activity, but lacking the ability to deliver a costim~ tory signal to T cells.
Such a peptide can be used to inhibit or block T cell proliferation and/or cytokine secretion in a subject. Alternatively, a peptide having both B7-2 binding activity and the ability to deliver a costimlll~tory signal to T cells is used to stimulate or enhance T cell proliferation and/or 30 cytokine secretion in a subject. Various modifications of the B7-2 protein to produce these and other functionally equivalent peptides are described in detail herein. The term "peptide"
as used herein, refers to peptides, proteins and polypeptides.
A peptide can be produced by modification of the amino acid sequence of the human B7-2 protein shown in Figure 8 (SEQ ID NO:2) or the murine B7-2 protein shown in Figure 35 14 (SEQ ID NO:23), such as a substitution, addition or deletion of an arnino acid residue which is not directly involved in the function of B7-2 (i.e., the ability of B7-2 to bind CTLA4 and/or CD28 and/or stim~ te or inhibit T cell costimlll~tion). Peptides of the invention are WO 95/03408 21 6 ~ 0 91 PCT/US94/084Z3 typically at least 20 amino acid residues in length, preferably at least 40 amino acid residues in length, and most preferably 60 amino acid residues in length. Peptides having B7-2 activity and including at least 80 amino acid residues in length, at least 100 arnino acid residues in length, or at least 200 or more amino acid residues in length are also within the - 5 scope of the invention. A pl~f~lled peptide includes an extracellular domain portion of the human B7-2 antigen (e.g., about amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2). Other preferred peptides have an amino acid sequence represented by a formula:

1 0 Xn~Y~Zm where Y is amino acid residues selected from the group consisting of: amino acid residues 55-68 of the sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 81 -89 of the sequence shown in Figure 8 (SEQ ID NO:2), amino acid residues 128-142 of the sequence shown in Figure 8 (SEQ ID NO:2), amino acid residues 160-169 of the sequence shown in Figure 8 (SEQ ID NO:2); arnino acid residues 188-200 of the sequence shown in Figure 8 (SEQ ID NO:2); and amino acid residues 269-282 of the sequence shown in Figure 8 (SEQ
ID NO:2). In the formula, Xn and Zm are additional amino acid residues linked to Y by an amide bond. Xn and Zm are amino acid residues selected from amino acids contiguous to Y
in the amino acid sequence shown in Figure 8 (SEQ ID NO:2). Xn is amino acid residues selected from amino acids contiguous to the amino terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2). Zm is amino acid residues selected from amino acids contiguous to the carboxy terminl~c of Y in the sequence shown in Figure 8 (SEQ ID NO:2). According to the formula, n is a number from 0 to 30 (n=0-30) and m is a number from 0 to 30 (m=0-30).
A particularly preferred peptide has an amino acid sequence represented by the formula Xn-Y~Zm~ where n=0 and m=0.
Another embodiment of the invention provides a subst~nti~lly pure ~ep~Lion of a peptide having an activity of a novel B lymphocyte antigen such as B7-2 or B7-3. Such a ,~l~dldLion is subst~nti~lly free of proteins and peptides with which the peptide naturally occurs in a cell or with which it naturally occurs when secreted by a cell.
The term "isolated" as used throughout this application refers to a nucleic acid, protein or peptide having an activity of a novel B Iymphocyte antigen, such as B7-2, subst~nti~lly free of cellular material or culture medium when produced by recombinant DNA techniques, or chemical precursors or other chemicals when chemic~lly synthesized.
An isolated nucleic acid is also free of sequences which naturally flank the nucleic acid (i.e., sequences located at the 5' and 3' ends of the nucleic acid) in the organism from which the nucleic acid is derived.

WO 9',/03408 ~ 9 1 PCT/US94/08423 These and other aspects of this invention are described in detail in the following - subsections.

k Isolation of Nucleic Acid From Cell T inec S Suitable cells for use in isolating nucleic acids encoding peptides having an activity of a novel B lymphocyte antigen include cells capable of producing mRNA coding for B
lymphocyte antigens (e.g., B7-1, B7-2, B7-3) and ~ Jpliately translating the mRNA into the corresponding protein. One source of mRNA is normal human splenic B cells, either resting or activated by treatment with an anti-immunoglobulin antibody or an anti-MHC class II antibody, or from subsets of neoplastic B cells. Expression ofthe human B7-2 antigen is cletect~hle in resting B cells and in activated B cells, with mRNA levels increasing 4-fold from resting levels following stim~ tion. Total cellular RNA can be obtained using standard techniques from resting or activated B cells during these intervals and utilized in the construction of a cDNA library.
In addition, various subsets of neoplastic B cells may express B7-2 and B7-3 and can ~ltern~tively serve as a source of the mRNA for construction of a cDNA library. For example, tumor cells isolated from patients with non-Hodgkins Iymphoma express B7-1 mRNA. B cells from nodular, poorly differenti~tt-d lymphoma (NPDL), diffuse large cell lymphoma (LCL) and Burkitt's lymphoma cell lines are also suitable sources of human B7-1 mRNA and, potentially B7-2 and B7-3 mRNA. Myelomas generally express B7-2, but not B7-1 mRNA, and, thus can provide a source of B7-2 mRNA. The Burkitt's Iymphoma cell line Raji is one source of B Iymphocyte antigen mRNA. Preferably, B7-2 mRNA is obtained from a population of both resting and activated normal human B cells. Activated B cells can be obtained by stim~ tion over a broad spectrum of time (e.g., from minlltes to days) with, for example, an anti-immllnoglobulin antibody or an anti-MCH class II antibody.

Tr. T~olation of mRNA ~nd Con~truction of cDNA L ibr~ry Total cellular mRNA can be isolated by a variety of techniques, e.g., by using the guanidinium-thiocyanate extraction procedure of Chirgwin et al., Biochemistry 18, 5294-5299 (1979). According to this method, Poly (A+) mRNA is prepared and purified for use in a cDNA library construction using oligo (dT) cellulose selection. cDNA is then synthe~i7~d from the poly(A+) RNA using oligo(dT) priming and reverse transcriptase. Moloney MLV
reverse transcriptase (available from Gibco/BRL, Bethesda, MD) or AMV reverse transcriptase (available from Seikagaku America, Inc., St. Petersburg, FL) are preferably employed.
Following reverse transcription, the mRNA/DNA hybrid molecule is converted to double stranded DNA using conventional techniques and incorporated into a suitable vector.

~WO 95/03408 21~ ~ ~ 9 ~ PCTIUS94/08423 The ex~ llents herein employed E. coli DNA polymerase I and ribonuclease H in the conversion to double stranded cDNA.
Cloning of the cDNAs can be accomplished using any of the conventional techniques for joining double stranded DNA with an appropriate vector. The use of synthetic adaptors is particularly preferred, since it alleviates the possibility of cleavage of the cDNA with restriction enzyme prior to cloning. Using this method, non-self complementary~ kin~efl adaptors are added to the DNA prior to ligation with the vector. Virtually any adaptor can be employed. As set forth in more detail in the examples below, non-self complementary BstXI
adaptors are preferably added to the cDNA for cloning, for ligation into a pCDM8 vector prepared for cloning by digestion with BstXI.
Eucaryotic cDNA can be e~ ,ssed when placed in the sense orientation in a vectorthat supplies an a~plo~l;ate eucaryotic promoter and origin of replication and other elements including enhancers, splice acceptors and/or donor sequences and polyadenylation signals.
The cDNAs of the present invention are placed in suitable vectors cont~ininp: a eucaryotic promoter, an origin of replication functional in E. coli, an SV40 origin of replication which allows growth in COS cells, and a cDNA insertion site. Suitable vectors include ~H3 (Seed and ~ruffo, Proc. Natl. Acad. Sci., 84:3365-3369 (1987)),7~H3m (Aruffo and Seed, Proc.
Natl. Acad. Sci., 84:8573-8577 (1987)), pCDM7 and pCDM8 (Seed, Nature, 329:840-841 (1987), with the pCDM8 vector being particularly ~lere,-ed (available commercially from Invitrogen, San Diego, CA).

TTT Tr~n~fection of Host Cells and Screenin~ for Novel B T,ymphocyte Activation Anti~ens The thus prepared cDNA library is then used to clone the gene of interest by t;x~essiorl cloning techniques. A basic expression cloning technique has been described by Seed and Aruffo, Proc. Natl. Acad. Sci. USA, 84:3365-3369 (1987) and Aruffo and Seed, Proc. Natl. Acad. Sci. USA, 84:8573-8577 (1987), although modifications to this technique may be n~cess~ry.
According to one embodiment, plasmid DNA is introduced into a simian COS cell line (Gluzman, Cell 23: 175 (1981)) by known methods of transfection (e.g., DEAE-Dextran) and allowed to replicate and express the cDNA inserts. The transfectants expressing B7-1 antigen are depleted with an anti-B7-1 monoclonal antibody (e.g., 133 and B1.1) and anti-murine IgG and IgM coated immunomagnetic beads. Transfectants expressing human B7-2 antigen can be positively selected by reacting the transfectants with the fusion proteins CTLA4Ig and CD28Ig, followed by panning with anti-human Ig antibody coated plates.
Although human CTLA4Ig and CD28Ig fusion proteins were used in the examples described herein, given the cross-species reactivity between B7-1 and, for example murine B7-1, it can be expected that other fusion proteins reactive with another cross-reactive species could be 2 ~

used. After p~nning! episomal DNA is recovered from the panned cells and transformed into a competent bacterial host, preferably E. coli. Plasmid DNA is subsequently reintroduced into COS cells and the cycle of expression and panning repeated at least two times. After the final cycle, plasmid DNA is prepared from individual colonies, transfected into COS cells and analyzed for expression of novel B Iymphocyte antigens by indirect immunofluorescence with, for example, CTLA4Ig and CD28Ig.

IV. Sequencin~ of Novel R T~y~hoc,vte ~nti~ens Plasmids are prepared from those clones which are strongly reactive with the CTLA4Ig and/or CD28Ig. These plasmids are then sequenced. Any of the conventional sequencing techniques suitable for sequencing tracts of DNA about 1.0 kb or larger can be employed.
As described in Example 4, a human B7-2 clone (clone29) was obtained cont~inin~ an insert of 1,120 base pairs with a single long open reading frame of 987 nucleotides and - 15 approximately 27 nucleotides of 3' noncoding sequences (Figure 8, SEQ ID NO: 1). The predicted amino acid sequence encoded by the open reading frame of the protein is shown below the nucleotide sequence in Figure 8. The encoded human B7-2 protein, is predicted to be 329 amino acid residues in length (SEQ ID NO:2). This protein sequence exhibits many features common to other type I Ig ~u~clr~llily membrane proteins. Protein translation is predicted to begin at the methionine codon (ATG, nucleotides 107 to 109) based on the DNA
homology in this region with the consensus eucaryotic translation initiation site (see Kozak, M. (1987) Nucl. ~4cids Res. 15:8125-8148). The amino terminll~ ofthe B7-2 protein (amino acids 1 to 23) has the characteristics of a secretory signal peptide with a predicted cleavage between the ~l~nin~s at positions 23 and 24 (von Heijne (1987) Nucl. Acids Res. 14:4683).
Processing at this site would result in a B7-2 membrane bound protein of 306 amino acids having an unmodified molecular weight of approximately 34 kDa. This protein would consist of an approximate extracellular Ig superfamily V and C like domains of from about amino acid residue 24 to 245, a hydrophobic tr~n~m~.mbrane domain of from about amino acid residue 246 to 268, and a long cytoplasmic domain of from about amino acid residue 269 to 329. The homologies to the Ig ~u~lr~llily are due to the two contiguous Ig-like ~lom~in~ in the extracellular region bound by the cysteines at positions 40 to 110 and 157 to 218. The extracellular domain also contains eight potential N-linked glycosylation sites and, like B7-1, is probably glycosylated. Glycosylation of the human B7-2 protein may increase the molecular weight to about 50-70 kDa. The cytoplasmic domain of human B7-2, while somewhat longer than B7-1, contains a common region of multiple cysteines followed by positively charged amino acids which pl~ulnably function as sign~ling or regulatory domains within an antigen-presenting cell (APC). Comparison of both the nucleotide and WO 9~/0340~ 21~ i~ O ~ 3 PCT/USg4/08423 amino acid sequences of the human B7-2 with the GenBank and EMBL ~l~t~b~ces yielded significant homology (about 26% amino acid sequence identity) with human B7-1. Since human B7-1, human B7-2 and murine B7-1 all bind to human CTLA4 and CD28, the homologous amino acids probably represent those necessary to comprise a CTLA4 or CD28 - 5 binding sequence. Æ. coli transfected with a vector cont~ining a cDNA insert encoding human B7-2 (clone 29) was deposited with the American Type Culture Collection (ATCC) on July 26, 1993 as Accession No. 69357.

V. Clo~in~ Novel P~ Lyrr~rhocyte ~nt~gens from Other M~mm~lian Species The present invention is not limited to human nucleic acid molecules and con~ plates that novel B lymphocyte antigen homologues from other m~mm~ n species that express B lymphocyte antigens can be cloned and sequenced using the techniques described herein. B lymphocyte antigens isolated for one species (e.g., hllm~n~) which exhibit cross-species reactivity may be used to modify T cell mediated immune responses in a different species (e.g., mice). Isolation of cDNA clones from other species can also be accomplished using human cDNA inserts, such as human B7-2 cDNA, as hybridizationprobes.
As described in Example 6, a murine B7-2 clone (mB7-2, clone 4) was obtained cont~ining an insert of 1,163 base pairs with a single long open reading frame of 927 nucleotides and approximately 126 nucleotides of 3' noncoding sequences (Figure 14, SEQ
ID NO:22). The predicted amino acid sequence encoded by the open reading frame of the protein is shown below the nucleotide sequence in ~igure 14. The encoded murine B7-2 protein, is predicted to be 309 amino acid residues in length (SEQ ID NO:23). This protein sequence exhibits many features common to other type I Ig superfamily membrane proteins.
Protein translation is predicted to begin at the methionine codon (ATG, nucleotides 111 to 113) based on the DNA homology in this region with the consensus eucaryotic translation initiation site (see Kozak, M. (1987) Nucl. Acids Res. 15:8125-8148). The amino terminus of the murine B7-2 protein (amino acids 1 to 23) has the characteristics of a secretory signal peptide with a predicted cleavage between the alanine at position 23 and the valine at position 24 (von Heijne (1987) Nucl. Acids Res. 14:4683). Processing at this site would result in a murine B7-2 membrane bound protein of 286 amino acids having an unmodified molecular weight of approximately 32 kDa. This protein would consist of an approximate extracellular Ig superfamily V and C like domains of from about amino acid residue 24 to 246, a hydrophobic transmembrane domain of from about amino acid residue 247 to 265, and a long cytoplasmic domain of from about amino acid residue 266 to 309. The homologies to the Ig superfarnily are due to the two contiguous Ig-like domains in the extracellular region bound by the cysteines at positions 40 to 110 and 157 to 216. The extracellular domain also WO 95/03408 ~ PCTIUSs4/08423 contains nine potential N-linked glycosylation sites and, like murine B7-1, is probably glycosylated. Glycosylation of the murine B7-2 protein may increase the molecular weight to about 50-70 kDa. The cytoplasmic domain of murine B7-2 contains a common region which has a cysteine followed by positively charged amino acids which presumably functions as S ~i~n~linf~ or regulatory domain within an APC. Comparison of both the nucleotide and amino acid sequences of murine B7-2 with the GenBank and EMBL cl~t~h~ees yieldedsignificant homology (about 26% amino acid sequence identity) with human and murine B7-1. Murine B7-2 exhibits about 50% identity and 67% similarity with its human homologue, hB7-2. E. coli (DH106/p3) transfected with a vector (plasmid pmBx4) cont~inin~ a cDNA
10 insert encoding murine B7-2 (clone 4) was deposited with the American Type Culture Collection (ATCC) on August 18, 1993 as Accession No. 69388.
Nucleic acids which encode novel B lymphocyte antigens from other species, such as the murine B7-2, can be used to generate either transgenic ~nim~l.c or "knock out" ~nim~l~
which, in turn, are useful in the development and screening of therapeutically useful reagents.
15 A tr~n~genic animal (e.g., a mouse) is an animal having cells that contain a transgene, which transgene was introduced into the animal or an ancestor of the animal at a prenatal, e.g., an embryonic stage. A transgene is a DNA which is integrated into the genome of a cell from which a tr~n~genic animal develops. In one embodiment, murine B7-2 cDNA or an al)~.o~.liate sequence thereof can be used to clone genomic B7-2 in accordance with 20 established techniques and the genomic sequences used to generate transgenic ~nim~ls that contain cells which express B7-2 protein. Methods for generating transgenic ~nim~
particularly ~nim~ls such as mice, have become conventional in the art and are described, for example, in U.S. Patent Nos. 4,736,866 and 4,870,009. Typically, particular cells would be targeted for B7-2 transgene incorporation with tissue specific enhancers, which could result 25 in T cell costimulation and enh~n~e~l T cell proliferation and autoimmllnity. Transgenic ~nim~l~ that include a copy of a B7-2 transgene introduced into the germ line of the animal at an embryonic stage can be used to examine the effect of increased B7 t;x~ ssion. Such ~nim~l~ can be used as tester ~nim~l~ for reagents thought to confer protection from, for example, autoilll,llulle disease. In accordance with this facet of the invention, an animal is 30 keated with the reagent and a reduced incidence of the fli~eZI'~Ç7 compared to untreated ~nim~l~ bearing the transgene, would indicate a potential therapeutic intervention for the dlsease.
Alternatively, the non-human homologues of B7-2 can be used to construct a B7-2 "knock out" animal which has a defective or altered B7-2 gene as a result of homologous 35 recombination between the endogenous B7-2 gene and altered B7-2 genomic DNA
introduced into an embryonic cell of the animal. For example, murine B7-2 cDNA can be used to clone genomic B7-2 in accordance with established techniques. A portion of the WO 95/03408 ~ ~ 6 7 0 91 PCT/US94/08423 . .

genomic B7-2 DNA (e.g., such as an exon which encodes an extracellular domain) can be deleted or replaced with another gene, such as a gene encoding a selectable marker which can be used to monitor integration. Typically, several kilobases of unaltered fl~nkin~ DNA (both at the 5' and 3' ends) are included in the vector (see e.g., Thomas, K.R. and Capecchi, M. R.
(1987) Cell ~1:503 for a description of homologous recombination vectors). The vector is introduced into an embryonic stem cell line (e.g., by electroporation) and cells in which the introduced DNA has homologously recombined with the endogçnous DNA are selected (see e.g., Li, E. et al. (1992) Cell 69:915). The selected cells are then injected into a blastocyst of an animal (e.g., a mouse) to form aggregation chimeras (see e.g., Bradley, A. inTeratocarcinomas and ~mbryonic Stem Cells: A Practical ~pproach, E.J. Robertson, ed.
(IRL, Oxford, 1987) pp. 113-152). A chimeric embryo can then be implanted into a suitable pseudopregnant female foster animal and the embryo brought to term to create a "knock out"
animal. Progeny harbouring the homologously recombined DNA in their germ cells can be identified by standard techniques and used to breed ~nim~l~ in which all cells of the animal contain the homologously recombined DNA. Knockout ~nim~l~ can be characterized for their ability to accept grafts, reject tumors and defend against infectious ~ e~es and can be used in the study of basic immllnobiology.

VI. Fxpression of B l.ym,~hocyte ~nt~el7~
Host cells transfected to express peptides having the activity of a novel B lymphocyte antigen are also within the scope of the invention. The host cell may be any procaryotic or eucaryotic cell. For exarnple, a peptide having B7-2 activity may be expressed in bacterial cells such as E. coli, insect cells (baculovirus), yeast, or m~mm~ n cells such as Chinese harnster ovary cells (CHO) and NS0 cells. Other suitable host cells may be found in Goeddel, (1990) supra or are known to those skilled in the art.
FOI exarnple, expression in eucaryotic cells such as m~mm~ n, yeast, or insect cells can lead to partial or complete glycosylation and/or formation of relevant inter- or intra-chain ~lixlllfi~le bonds of recombinant protein. Exarnples of vectors for e~pression in yeast 5. cerivisae include pYepSec l (Baldari. et ~L, (1987) Embo J. 6:229-234), pMFa (Kurjan and Herskowitz, (1982) Cell 30:933-943), pJRY88 (Schultz et al., (1987) Gene 54:113-123), and pYES2 (Invitrogen Corporation, San Diego, CA). Baculovirus vectors available forexpression of proteins in cultured insect cells (SF 9 cells) include the pAc series (Smith et al., (1983) Mol. Cell Biol. 3:2156-2165) and the pVL series (Lucklow, V.A., and Surnrners, M.D., (198g) Virology 170:31-39). Generally, COS cells (Gluzman, Y., (1981) Cell ;~:175-182) are used in conjunction with such vectors as pCDM8 (Seed, B., (1987) Nature ~2:840) for transient arnplification/e~les~ion in m~mm?~ n cells, while CHO (dhfr~ Chinese ~arnster Ovary~ cells are used with vectors such as pMT2PC (~llfm~n et ~L (1987), WO 95/03408 ; PCT/US94/08423 7 ~ 24-EMBO J. 6:187-195) for stable amplification/~x~ ion in m~mm~ n cells. A preferred cell line for production of recombinant protein is the NS0 myeloma cell line available from the ECACC (catalog #85110503) and described in Galfre, G. and Milstein, C. ((1981) Methods in Enzymology Z~(13):3-46; and Preparation of Monoclonal Antibodies: Strategies 5 and Procedures, Academic Press, N.Y., N.Y). Vector DNA can be introduced into m~mm~ n cells via conventional techniques such as calcium phosphate or calcium chloride co-precipitation, DEAE-dextran-mediated transfection, lipofectin, or electroporation.
Suitable methods for transforming host cells can be found in Sambrook et ~L (Molecular Cloning: A Laboratory Manual, 2nd Edition, Cold Spring Harbor Laboratory press (1989)), 10 and other laboratory textbooks. When used in m~mm~ n cells, the expression vector's control functions are often provided by viral material. For example, commonly used promoters are derived from polyoma, Adenovirus 2, cytomegalovirus and most frequently, Simian Virus 40.
It is known that a small faction of cells (about I out of 105) typically integrate DN~
15 into their genomes. In order to identify these integrants, a gene that contains a selectable marker (i.e., resistance to antibiotics) is generally introduced into the host cells along with the gene of interest. Preferred selectable markers include those which confer resistance to drugs, such as G418, hygromycin and methollc;xal~. Selectable markers may be introduced on the same plasmid as the gene of interest or may be introduced on a separate plasmid. Cells 20 cont~ining the gene of interest can be identified by drug selection; cells that have incorporated the selectable marker gene will survive, while the other cells die. The surviving cells can then be screened for production of novel B lymphocyte antigens by cell surface staining with ligands to the B cell antigens (e.g., CTLA4Ig and CD28Ig). Alternatively, the protein can be metabolically radiolabeled with a labeled amino acid and immlln~precipitated 25 from cell s~c;~ t with an anti-B lymphocyte antigen monoclonal antibody or a fusion protein such as CTLA4Ig or CD28Ig.
Expression in procaryotes is most often carried out in E coli with vectors cont~ining constitutive or inducible promotors directing the e~res~ion of either fusion or non-fusion proteins. Fusion vectors add a number of amino acids usually to the amino terminus of the 30 expressed target gene. Such fusion vectors typically serve three purposes: 1) to increase sion of recombinant protein; 2) to increase the solubility of the target recombinant protein; and 3) to aid in the purification of the target recombinant protein by acting as a ligand in affinity purification. Often, in fusion expression vectors, a proteolytic cleavage site is introduced at the junction of the fusion moiety and the target recombinant protein to enable 35 separation of the target recombinant protein from the fusion moiety subsequent to purification of the fusion protein. Such enzymes, and their cognate recognition sequences, include Factor Xa, thrombin and enterokinase. Typical fusion expression vectors include pGEX (Amrad WO 95/03408 2 I G ~ O 91 PCT/US94/08423 Corp., Melbourne, Australia), pMAL (New England Biolabs, Beverly, MA) and pRIT5 (Pharrnacia, Piscataway, NJ) which fuse glutathione S-tranferase, maltose E binding protein, or protein A, respectively, to the target recombinant protein.
E coli ~x~ ssion systems include the inducible ~x~.~ssion vectors pTrc (Amann et 5 ~L, (1988) Gene 69:301 -315) and pET 11 (Studier et ~L, Gene Expression Technology.
Methods in Enzymology 185, Academic Press, San Diego, California (1990) 60-89;
commercially available from Novagen). In the pTrc vector system, the inserted gene is expressed with a pelB signal sequence by host RNA polymerase transcription from a hybrid trp-lac fusion promoter. After induction, the recombinant protein can be purified from the periplasmic fraction. In the pET 11 vector system, the target gene is expressed as non-fusion protein by transcription from the T7 gnlO-lac 0 fusion promoter mediated by a coexpressed viral RNA polymerase (T7 gnl ). This viral polymerase is supplied by host E. coli strains BL21 (DE3) or HMS 174(DE3) from a resident ~ prophage harboring a T7 gnl under the transcriptional control of the lacUV S promoter. In this system, the recombinant protein can be purified from inclusion bodies in a denatured form and, if desired, renatured by step gradient dialysis to remove denaturants.
One strategy to maximize recombinant B7-2 expression in E. coli is to express the protein in a host bacteria ~,vith an impaired capacity to proteolytically cleave the recombinant protein (Gotte~m~n, S., Gene Expression Technology: Methods in En~ymology 18$, ~c~lemic Press, San Diego, California (1990) 119-128). Another strategy would be to alter the nucleic acid sequence of the B7-2 gene or other DNA to be inserted into an expression vector so that the individual codons for each amino acid would be those preferentially utilized in highly expressed E. coli proteins (Wada et ~LL, (1992) Nuc. ~cids Res. ~Q:2111-2118).
Such alteration of nucleic acid sequences of the invention could be carried out by standard DNA synthesis techniques.
Novel B lymphocyte antigens and portions thereof, expressed in m~mm~ n cells or otherwise, can be purified according to standard procedures of the art, including ammonium sulfate ~lecipiL~Iion, fractionation column chromatography (e.g. ion exchange, gel filtration, electrophoresis, affinity chromatography, etc.) and ultimately, crystallization (see generally, "Enzyme Purification and Related Techniques", Methods in Enzymolo~y, 22:233-577 (1971)). Orlce purified, partially or to homogeneity, the recombinantly produced B
lymphocyte antigens or portions thereof can be utilized in compositions suitable for ph~rm~ceutical ~fimini.~tration as described in detail herein.

WO 95/03408 ~ ~ 6 ~ ~ 91 PCT/US94/08423 VIT. Motlifications of Nuçleic Acid and Amino Acid Sequences of the Invention ~n~l Assays for B7 J ~y~hocvte Anti~en Activity It will be appreciated by those skilled in the art that other nucleic acids encoding peptides having the activity of a novel B lymphocyte antigen can be isolated by the above 5 process. Different cell lines can be expected to yield DNA molecules having different sequences of bases. Additionally, variations may exist due to genetic polymorphisms or cell-mediated modifications of the genetic material. Furthermore, the DNA sequence of a B
lymphocyte antigen can be modified by genetic techniques to produce proteins or peptidçs with altered amino acid sequences. Such sequences are considered within the scope of the 10 present invention, where the expressed peptide is capable of either inducing or inhibiting activated T cell mediated immune responses and imml-ne function.
A number of processes can be used to generate equivalents or fragments of an isolated DNA sequence. Small subregions or fr~gment~ of the nucleic acid encoding the B7-2 protein, for example 1-30 bases in length, can be prepared by standard, synthetic organic 15 chemical means. The technique is also useful for ~ .~dtion of antisense oligonucleotides and primers for use in the generation of larger synthetic fr~gment~ of B7-2 DNA.Larger subregions or fragments of the genes encoding B lymphocyte antigens can be expressed as peptides by syntht?~i7inE the relevant piece of DNA using the polymerase chain reaction (PCR) (Sambrook, Fritsch and ~ni~ti~, 2 Molecular Cloning; A Laboratory20 Manual, Cold Spring Harbor, N.Y., (1989)), and lig~ting the thus obtained DNA into an a~,v~l;ate expression vector. Using PCR, specific sequences ofthe cloned double stranded DNA are generated. cloned into an expression vector, and then assayed for CTLA4/CD28 binding activity. For example, to express a secreted (soluble) form of the human B7-2 protein, using PCR, a DNA can be synthesized which does not encode the transmembrane 25 and cytoplasmic regions of the protein. This DNA molecule can be ligated into an a~ ,;ate expression vector and introduced into a host cell such as CHO, where the B7-2 protein fragment is synthesized and secreted. The B7-2 protein fragment can then readily be obtained from the culture media.
In another embodiment, mutations can be introduced into a DNA by any one of a 30 number of methods, including those for producing simple deletions or insertions, systematic deletions, insertions or substitutions of clusters of bases or substitutions of single bases, to generate variants or modified equivalents of B lymphocyte antigen DNA. For example, changes in the human B7-2 cDNA sequence shown in Figure 8 (SEQ ID NO:1) or murine B7-2 cDNA sequence shown in Figure 14 (SEQ ID NO:22) such as amino acid substitutions 35 or deletions are preferably obtained by site-directed mutagenesis. Site directed mutagenesis systems are well known in the art. Protocols and reagents can be obtained commercially from Amersham Tntem~tional PLC, Amersham, U.K.

3 ~ 1 6 7 0 91 PCTIUS94/08423 Peptides having an activity of a novel B lymphocyte antigen, i.e., the ability to bind to the natural ligand(s) of a B lymphocyte antigen on T cells and either stim~ te (amplify) or inhibit (block) activated T cell mediated immlme responses, as evidenced by, for example, cytokine production and/or T cell proliferation by T cells that have received a primary 5 activation signal are considered within the scope of the invention. More specifically, peptides that bind to T lymphocytes, for example CD28+ cells, may be capable of delivering a costim~ tory signal to the T lymphocytes, which, when transmitted in the presence of antigen and class II MHC, or other material capable of tr~n~mit~in~ a primary signal to the T
cell, results in activation of cytokine genes within the T cell. Alternatively, such a peptide 10 can be used in conjunction with class I MHC to thereby activate CD8+ cytolytic T cells. In addition, soluble, monomeric forms of the B7-2 protein, may retain the ability to bind to their natural ligand(s) on CD28+ T cells but, perhaps because of insufficient cross-linking with the ligand, fail to deliver the secondary signal essential for enhanced cytokine production and cell division. Such peptides, which provide a means to induce a state of anergy or tolerance in the 15 cells, are also considered within the scope of the invention.
Screening the peptides for those which retain a characteristic B lymphocyte antigen activity as described herein can be accomplished using one or more of several different assays. For example, the peptides can be screened for specific reactivity with an anti-B7-2 monoclonal antibody reactive with cell surface B7-2 or with a fusion protein, such as 20 CTLA4Ig or CD28Ig. Specifically, appropriate cells, such as COS cells, can be transfected with a B7-2 DNA encoding a peptide and then analyzed for cell surface phenotype by indirect immunofluorescence and flow cytometry to determine whether the peptide has B7-2 activity.
Cell surface e~ es~ion of the transfected cells is evaluated using a monoclonal antibody specifically reactive with cell surface B7-2 or with a CTLA4Ig or CD28Ig fusion protein.
25 Production of secreted forms of B7-2 is evaluated using anti-B7-2 monoclonal antibody or CTLA4Ig or CD28 fusion protein for immllnoprecipitation.
Other, more ~,le~lled, assays take advantage of the functional characteristics of the B7-2 antigen. As previously set forth, the ability of T cells to synthesi7~ cytokines depends not only on occllp~nGy or cross-linking of the T cell receptor for antigen (the "prima~y 30 activation signal" provided by, for example anti-CD3, or phorbol ester to produce an "activated T cell"), but also on the induction of a costimnl~tt ry signal, in this case, by interaction with a B lymphocyte antigen, such as B7-2, B7-1 or B7-3. The binding of B7-2 to its natural ligand(s) on, for example, CD28+ T cells, has the effect of transmitting a signal to the T cell that induces the production of increased levels of cytokines, particularly of 35 interleukin-2, which in turn ~tim~ tt-s the proliferation of the T lymphocytes. Other assays for B7-2 function thus involve assaying for the synthesis of cytokines, such as interleukin-2, 2; 1 6 ~ 28-interleukin-4 or other known or unknown novel cytokines? and/or assaying for T cell proliferation by CD28+ T cells which have received a primary activation signal.
In vitro, T cells can be provided with a first or primary activation signal by anti-T3 monoclonal antibody (e.g. anti-CD3) or phorbol ester or, more preferably, by antigen in 5 association with class II MHC. T cells which have received a primary activation signal are referred to herein as activated T cells. B7-2 function is assayed by adding a source of B7-2 (e.g., cells expressing a peptide having B7-2 activity or a secreted form of B7-2) and a ~1;111~.~ activation signal such as antigen in association with Class II MHC to a T cell culture and assaying the culture supernatant for interleukin-2, garnma interferon, or other known or 10 unknown cytokine. For example, any one of several conventional assays for interleukin-2 can be employed, such as the assay described in Proc. Natl. Acad Sci. USA, 86: 1333 (1989) the pertinent portions of which are incorporated herein by reference. A kit for an assay for the production of interferon is also available from Genzyme Corporation (Cambridge, MA.).
T cell proliferation can also be measured as described in the Examples below. Peptides that 15 retain the characteristics of the B7-2 antigen as described herein may result in increased per cell production of cytokines, such as IL-2, by T cells and may also result in enhanced T cell proliferation when colllp~d to a negative control in which a costim~ tory signal is lacking.
The same basic functional assays can also be used to screen for peptides having B7-2 activity, but which lack the ability to deliver a costimulatory signal, but in the case of such 20 peptides, addition of the B7-2 protein will not result in a marked increase in proliferation or cytokine secretion by the T cells. The ability of such proteins to inhibit or completely block the normal B7-2 costim~ tory signal and induce a state of anergy can be determined using subsequent attempts at stimlll~tion of the T cells with antigen pres~nting cells that express cell surface B7-2 and present antigen. If the T cells are unresponsive to the subsequent 25 activation attempts, as determined by IL-2 synthesis and T cell proliferation~ a state of anergy has been in~ ce~l See, e.g., Gimmi, C.D. et al. (1993) Proc. Natl. Acad. Sci. USA ~Q, 6586-6590, and Schwartz (1990) Science, 248, 1349-1356, for assay systems that can used as the basis for an assay in accordance with the present invention.
It is possible to modify the structure of a peptide having the activity of a novel B
30 lymphocyte antigen for such purposes as increasing solubility, enhancing therapeutic or prophylactic efficacy, or stability (e.g., shelf life ex vivo and resistance to proteolytic degradation in vivo). Such modified peptides are considered functional equivalents of the B
lymphocyte antigens as defined herein. For example, a peptide having B7-2 activity can be modified so that it m~int~in~ the ability to co-stimulate T cell proliferation and/or produce 35 cytokines. Those residues shown to be ess~nti~l to interact with the CTLA4/CD28 receptors on T cells can be modified by replacing the essenti~l amino acid with another, preferably similar amino acid residue (a conse, v~liv~ substitution) whose presence is shown to enhance, WO 95/03408 ~ ~ fi ~ ~ ~1 PCT/US94/08423 fiimini~h, but not elimin~te or not effect receptor interaction. In addition. those amino acid residues which are not essential for receptor interaction can be modified by being replaced by another amino acid whose incorporation may enhance, ~imini~h, or not effect reactivity.
Another example of modification of a peptide having the activity of a novel B
- S lymphocyte antigen is substitution of cysteine residues preferably with ~l~nine, serine, threonine, leucine or glutamic acid residues to minimi7~ dimerization via ~ llfide linkages.
In addition, amino acid side chains of a peptide having B7-2 activity can be chemically modified. Another modification is cyclization of the peptide.
In order to enhance stability and/or reactivity, peptides having B7-2 activity can be modified to incorporate one or more polymorphisms in the amino acid sequence of the antigen resnlting from any natural allelic variation. Additionally, D-arnino acids, non-natural amino acids, or non-amino acid analogs can be substituted or added to produce a modified protein within the scope of this invention. Furthermore, the peptides can be modified using polyethylene glycol (PEG) according to the method of A. Sehon and co-workers (Wie et ~L, supra) to produce a peptide conjugated with PEG. In addition, PEG can be added during chemical synthesis of the peptide. Other modifications of the peptides include reduction/alkylation (Tarr in: Methods of Protein Microcharacterization, J. E. Silver ed., Hurnana Press, Clifton NJ 155-194 (1986)); acylation (Tarr, supra); chemical coupling to an a~plo~l;ate carrier (Mishell and Shiigi, eds, ~electe~Methods in Cellular Immunology, WH
Freeman, San Francisco, CA (1980), U.S. Patent 4,939,239; or mild formalin tre~tment (Marsh (1971), lnt. Arch. of Aller~ andAppl. Immunol. 41:199-215).
To facilitate purification and potentially increase solubility of a peptide, it is possible to add an amino acid fusion moiety to the protein backbone. For example, hexa-hi~ticiine can be added to the peptide for purification by immobilized metal ion affinity chromatography (Hochuli, E. et ~L, (1988) Bio/Technology 6:1321-1325). In addition, to facilitate isolation of a B lymphocyte antigen free of irrelevant sequences, specific endoprotease cleavage sites can be introduced between the sequences of a fusion moiety and the peptide. It may be necessary to increase the solubility of a peptide by adding functional groups to the peptide, or by omitting hydrophobic regions of the peptide.
VIT. Uses of Nucleic Acid Sequçnces Fnco~ B T ~n~l~hocyte Anti~e~ ~nd Peptides Hav;~ B7-2 Activity A. MolecularProbes The nucleic acids of this invention are useful diagnostically, for tracking the progress of tli~e~e, by measuring the activation status of B lymphocytes in biological samples or for assaying the effect of a molecule on the ~ esssion of a B Iymphocyte antigen (e.g., ~letecting cellular mRNA levels). In accordance with these diagnostic assays, the nucleic acid sequences are labeled with a detectable marker, e.g., a radioactive, fluorescent, or biotinylated marker and used in a conventional dot blot or Northern hybridization procedure to probe mRNA molecules of total or poly(A+) RNAs from a biological sample.
.
R. Antibody Production The peptides and fusion proteins produced from the nucleic acid molecules of thepresent invention can also be used to produce antibodies specifically reactive with B
lymphocyte antigens. For example, by using a full-length B7-2 protein, or a peptide fragment thereof, having an amino acid sequence based on the predicted amino acid sequence of B7-2, anti-protein/anti-peptide polyclonal antisera or monoclonal antibodies can be made using standard methods. A m~mm~l, (e.g., a mouse, h~met~r, or rabbit) can be immunized with an immlln~genic form of the protein or peptide which elicits an antibody response in the m~mm~l The immunogen can be, for example, a recombinant B7-2 protein, or fragment thereof, a synthetic peptide fragment or a cell that expresses a B lymphocyte antigen on its surface. The cell can be for example, a splenic B cell or a cell transfected with a nucleic acid encoding a B Iymphocyte antigen of the invention (e.g., a B7-2 cDNA) such that the B
lymphocyte antigen is expressed on the cell surface. The immllnogen can be modified to increase its immllnogenicity. For example, techniques for conferring immunogenicity on a peptide include conjugation to carriers or other techniques well known in the art. For example, the peptide can be ~lminietered in the presence of adjuvant. The progress of imml~ni7~tion can be monitored by detection of antibody titers in plasma or serum. Standard ELISA or other immllno~ee~y can be used with the immunogen as antigen to assess the levels of antibodies.
Following immllni7~tion, antisera can be obtained and, if desired, polyclonal antibodies isolated from the sera. To produce monoclonal antibodies, antibody producing cells (lymphocytes) can be harvested from an immlmi7e-1 animal and fused with myeloma cells by standard somatic cell fusion procedures thus immortalizing these cells and yielding hybridoma cells. Such techniques are well known in the art. For example, the hybridoma technique originally developed by Kohler and Milstein (Nature (1975) ~:495-497) as well as other techniques such as the human B-cell hybridoma technique (Kozbar et al.~ Immunol.
Today (19~3) 4:72), the EBV-hybridoma technique to produce human monoclonal antibodies (Cole et al. Monoclonal Antibodies in Cancer Therapy (1985) (Allen R. Bliss, Inc., pages 77-96), and screening of combinatorial antibody libraries (Huse et al., Science ( l 989) ~:1275).
Hybridoma cells can be screened immunochemically for production of antibodies specifically reactive with the peptide and monoclonal antibodies isolated.

WO 95/03401~ 21 ~ 1 PCT/US94/08423 The term antibody as used herein is intended to include fragments thereof which are also specifically reactive with a peptide having the activity of a novel B Iymphocyte antigen or fusion protein as described herein. Antibodies can be fragmented using conventional techniques and the fragments screened for utility in the same manner as described above for 5 whole antibodies. For exarnple, F(ab')2 fragments can be generated by treating antibody with pepsin. The resulting F(ab')2 fragment can be treated to reduce tli~ulfi~le bridges to produce Fab' frAgment~. The antibody of the present invention is further intended to include bispecific and chimeric molecules having an anti-B Iymphocyte antigen (i.e., B7-2, B7-3) portion.
Particularly preferred antibodies are anti-human B7-2 monoclonal antibodies produced by hybridomas HA3.1 F9, HA5.2B7 and HF2.3D 1. The p~cl)al~Lion and characterization of these antibodies is described in detail in Example 8. Monoclonal antibody HA3.1F9 was determined to be ofthe IgG1 isotype; monoclonal antibody HA5.2B7 wasdetermined to be of the IgG2b isotype; and monoclonal anibody HF2.3D I was determined to be of the IgG2a isotype. Hybidoma cells were deposited with the American Type Culture Collection, which meets the requirements of the Budapest Treaty, on July 19, 1994 as ATCC
AccessionNo. (hybridomaHA3.1F9),ATCCAccessionNo. (HA5.2B7)and ATCC Accession No. (HF2.3Dl).
When antibodies produced in non-human subjects are used therapeutically in h~lm~n~, they are recognized to varying degrees as foreign and an immune response may be generated in the patient. One approach for minimi7ing or elimin~ting this problem, which is preferable to general immlm~suppression~ is to produce chimeric antibody derivatives, i.e., antibody molecules that combine a non-human animal variable region and a human constant region.
Chimeric antibody molecules can include, for example, the antigen binding domain from an antibody of a mouse, rat, or other species, with human constant regions. A variety of approaches for m~king chimeric antibodies have been described and can be used to make chimeric antibodies cont~ining the imml-noglobulin variable region which recognizes the gene product of the novel B Iymphocyte antigens of the invention. See, for example, Morrison et al., Proc. Natl. Acad. Sci. U.S.A. 81 :6851 (1985); Takeda et al., Nature 314:452 (1985), Cabilly et al., U.S. Patent No. 4,816,567; Boss et al., U.S. Patent No. 4,816,397;
Tanaguchi et al., European Patent Publication EP171496; European Patent Publication 0173494, United Kingdom Patent GB 2177096B. It is expected that such chimeric antibodies would be less immllnogenic in a human subject than the corresponding non-chimeric antibody.
For human therapeutic purposes, the monoclonal or chimeric antibodies specifically reactive with a peptide having the activity of a B lymphocyte antigen as described herein can be further hl-m~ni7e~1 by producing human variable region chimeras, in which parts of the WO 95/03408 . PCT/US94/08423 ~1~7~9~ ~

variable regions, especially the conserved framework regions of the antigen-binding domain, are of human origin and only the hypervariable regions are of non-human origin. General reviews of "hllm~ni7to~1" chimeric antibodies are provided by Morrison, S. L. (1985) Science ~2:1202-1207 and by Oi et al. (1986) BioTechniques _:214. Such altered immunoglobulin molecules may be made by any of several techniques known in the art? (e.g., Teng et al., Proc. Natl. Acad. Sci. U.S.A., 80:7308-7312 (1983); Kozbor et al., Immunology Today, ~:7279 (1983); Olsson et al., Meth. Enzymol., 92:3-16 (1982)), and are preferably made according to the te~chin~ of PCT Publication WO92/06193 or EP 0239400. Hllm~ni7~d antibodies can be commercially produced by, for example, Scotgen Limited, 2 Holly Road, Twickenham, Middlesex, Great Britain. Suitable "hllm~ni7~1" antibodies can be alternatively produced by CDR or CEA substitution (see U.S. Patent 5,225,539 to Winter;
Jones et al. (1986) Nature ~1:552-525; Verhoeyan et al. (1988) Science 239:1534; and Beidler et al. (1988) J. Immunol. 141 :4053-4060). Hllm~ni7ecl antibodies which have reduced immun~genicity are preferred for immlln~therapy in human subjects.
Immunotherapy with a hl-m~ni7~-1 antibody will likely reduce the necessity for any concomitant imm--nosuppression and may result in increased long term effectiveness for the tre~tment of chronic disease situations or situations requiring repeated antibody tre~tment.~.
As an alterntive to hl-m~ni7ing a monoclonal antibody from a mouse or other species, a human monoclonal antibody directed against a human protein can be generated. Transgenic mice carrying human antibody repertoires have been created which can be immunized with a human B lymphocyte antigen, such as B7-2. Splenocytes from these immunized transgenic mice can then be used to create hybridomas that secrete human monoclonal antibodies specifically reactive with a human B lymphocyte antigen (see, e.g., Wood et al. PCT
publication WO 91/00906, Kucherlapati et al. PCT publication WO 91/10741; Lonberg et al.
PCT publication WO 92/03918; Kay et al. PCT publication 92/03917; Lonberg, N. et al.
(1994) Nature ~:856-859; Green, L.L. et al. (1994) Nature Genet. 1:13-21; Morrison, S.L.
et al. (1994) Proc. Natl. Acad. Sci. USA 81 :6851-6855; Bruggeman et al. (1993) Year Immunol 1:33-40; Tuaillon et al. (1993) PNAS 90:3720-3724; and Bruggeman et al. (1991) ~ur JImmunol ~1:1323-1326).
Monoclonal antibody compositions of the invention can also be produced by other methods well known to those skilled in the art of recombinant DNA technology. Analternative method, referred to as the "combinatorial antibody display" method, has been developed to identify and isolate antibody fragments having a particular antigen specificity, and can be utilized to produce monoclonal antibodies that bind a B lymphocyte antigen of the invention (for descriptions of combinatorial antibody display see e.g., Sastry et al. ~1989) PNAS~:5728; Huse et al. (1989) Science 246:1275; and Orlandi et al. (1989) PNAS
86:3833). After immunizing an animal with a B lymphocyte antigen, the antibody repertoire ~WO ~5/D3408 33 PCTIU594/08423 of the resulting B-cell pool is cloned. Methods are generally known for directly obtaining the DNA sequence of the variable regions of a diverse population of immunoglobulin molecules by using a mixture of oligomer primers and PCR. For instance, mixed oligonucleotide primers corresponding to the S' leader (signal peptide) sequences and/or framework 1 (FRl ) 5 sequences, as well as primer to a conserved 3' constant region primer can be used for PCR
amplification of the heavy and light chain variable regions from a number of murine antibodies (Larrick et al. (1991) Biotechniques 11:152-156). A similar strategy can also been used to amplify human heavy and light chain variable regions from human antibodies (Larrick et al. (1991) Methods: Companion fo Methods in Enzymology ~: 106- 110).In an illustrative embodiment, RNA is isolated from activated B cells of, for example, peripheral blood cells, bone marrow, or spleen ~lel)aldlions, using standard protocols (e.g., U.S. Patent No. 4,683,`202; Orlandi, et al. PNAS (1989) 86:3833-3837; Sastry et al., PNAS
(1989) 86:5728-5732; and Huse et al. (1989) Science ~:1275-1281.) First-strand cDNA is synthesi7P~l using primers specific for the constant region of the heavy chain(s) and each of 15 the K and ~ light chains, as well as primers for the signal sequence. Using variable region PCR primers, the variable regions of both heavy and light chains are amplified, each alone or in combinantion, and ligated into al)~ro~l;ate vectors for further manipulation in generating the display packages. Oligonucleotide primers useful in amplification protocols may be unique or degenerate or incorporate inosine at degenerate positions. Restriction endonuclease 20 recognition sequences may also be incorporated into the primers to allow for the cloning of the amplified fragment into a vector in a predetermined reading frame for expression.
The V-gene library cloned from the il,,,..l..li7~tion-derived antibody repertoire can be expressed by a population of display packages, preferably derived from filamentous phage, to form an antibody display library. Ideally, the display package comprises a system that allows the sampling of very large diverse antibody display libraries, rapid sorting after each affinity separation round, and easy isolation of the antibody gene from purified display packages. In addition to cornmercially available kits for gen~r~tin?~ phage display libraries (e.g., the Pharmacia Recombinant Phage Antibody System, catalog no. 27-9400-01; and the Stratagene Sur~ZAPTM phage display kit, catalog no. 240612), exarnples of methods and reagents particularly arnenable for use in generating a diverse antibody display library can be found in, for exarnple, Ladner et al. U.S. Patent No. 5,223,409; Kang et al. International Publication No. WO 92/18619; Dower et al. International Publication No. WO 91/17271; Winter et al.
International Publication WO 92/20791; Markland et al. International Publication No. WO
92/15679; Breitling et al. International Publication WO 93/01288, McCafferty et al.
International Publication No. WO 92/01047; Garrard et al. Tntetn~tional Publication No. WO
92/09690; Ladner et al. International Publication No. WO 90/02809; Fuchs et al. (1991) Bio/Technology 2: 1370-1372; Hay et al. (1992) Hum Antibod Hybridomas 3 :81 -85; Huse et WO 95/03408 PCT/US94l08423 2 ~ 9 ~ -34-al. (1989) Science ~:1275-1281; Griffths et al. (1993) EMBO J12:725-734; Hawkins et al.
(1992) JMol Biol ~:889-896; Clackson et al. (1991) Nature ~:624-628; Gram et al.(1992) PNAS ~2:3576-3580; Garrad et al. (1991) Bio/Technology 2:1373-1377; Hoogenboom et al. (1991) Nuc Acid Res 12:4133-4137; and Barbas et al. (1991) PN,45 88:7978-7982.
In certain embodiments, the V region domains of heavy and light chains can be e~ essed on the same polypeptide, joined by a flexible linker to form a single-chain Fv fragment, and the scFV gene subsequently cloned into the desired expression vector or phage genome. As generally described in McCafferty et al., Nature (1990) ~:552-554, complete VH and VL domains of an antibody, joined by a flexible (Gly4-Ser)3 linker can be used to produce a single chain antibody which can render the display package separable based on antigen affinity. Isolated scFV antibodies immunoreactive with a peptide having activity of a - B lymphocyte antigen can subsequently be formulated into a pharm~entical p~ ua,dlion for use in the subject method.
Once displayed on the surface of a display package (e.g., filamentous phage), the antibody library is screened with a B lymphocyte antigen protein, or peptide fragment thereof, to identify and isolate packages that express an antibody having specificity for the B
lymphocyte antigen. Nucleic acid encoding the selected antibody can be recovered from the display package (e.g., from the phage genome) and subcloned into other exples~ion vectors by standard recombinant DNA techniques.
The antibodies of the current invention can be used therapeutically to inhibit T cell activation through blocking receptor:ligand interactions necess~ry for costimulation of the T
cell. These so-called "blocking antibodies" can be identified by their ability to inhibit T cell proliferation and/or cytokine production when added to an in vitro costimulation assay as described herein. The ability of blocking antibodies to inhibit T cell functions may result in immunosuppression and/or tolerance when these antibodies are ~lmini~tered in vivo.

C. Protein Purification The polyclonal or monoclonal antibodies of the current invention, such as an antibody specifically reactive with a recombinant or synthetic peptide having B7-2 activity or B7-3 activity can also be used to isolate the native B lymphocyte antigen from cells. For example, antibodies reactive with the peptide can be used to isolate the naturally-occurring or native form of B7-2 from activated B lymphocytes by immllno~ffinity chromatography. In addition, the native form of B7-3 can be isolated from B cells by immunoaffinity chromatography with monoclonal antibody BB-l.

WO 95/03408 2 ~ ~ 7 ~ 91 PCT/US94/08423 D. Other Therapeutic Reagents The nucleic acid sequences and novel B Iymphocyte antigens described herein can be used in the development of therapeutic reagents having the ability to either upregulate (e.g., amplify) or downregulate (e.g., suppress or tolerize) T cell mediated immune responses. For 5 example, peptides having B7-2 activity, including soluble, monomeric forms of the B7-2 antigen or a B7-2 fusion protein, e.g., B7-2Ig, and anti-B7-2 antibodies that fail to deliver a costimulatory signal to T cells that have received a primary activation signal, can be used to block the B7-2 ligand(s) on T cells and thereby provide a specific means by which to cause immunosuppression and/or induce tolerance in a subject. Such blocking or inhibitory forms 10 of B lymphocyte antigens and fusion proteins and blocking antibodies can be identified by their ability to inhibit T cell proliferation and/or cytokine production when added to an in vitro costimulation assay as previously described herein. In contrast to the monomeric form, stimulatory forms of B7-2, such as an intact cell surface B7-2, retain the ability to transmit the costimulatory signal to the T cells, resulting in an increased secretion of cytokines when 15 compared to activated T cells that have not received the secondary signal.
In addition, fusion proteins compri~ing a first peptide having an activity of B7-2 fused to a second peptide having an activity of another B lymphocyte antigen (e.g., B7-1) can be used to modify T cell mediated immllne responses. ~ ely, two separate peptides having an activity of B lymphocyte antigens, for example, B7-2 and B7- 1, or a combination 20 of blocking antibodies (e.g., anti-B7-2 and anti-B7- 1 monoclonal antibodies) can be combined as a single composition or ~(lmini~tered st;~ ely (simultaneously or sequentially), to upregulate or downregulate T cell mediated immune responses in a subject.
Furthermore, a therapeutically active amount of one or more peptides having B7-2 activity and or B7-1 activity can be used in conjunction with other immunomod-~l~tin~ reagents to 25 influence immune responses. Exarnples of other immllnomo~ ting reagents include blocking antibodies, e.g., against CD28 or CTLA4, against other T cell markers or against cytokines, fusion proteins, e.g., CTLA4Ig, or immunosuppressive drugs, e.g., cyclosporine A
or FK506.
The peptides produced from the nucleic acid molecules of the present invention may 30 also be useful in the construction of therapeutic agents which block T cell function by destruction of the T cell. For example, as described, secreted forms of a B lymphocyte antigen can be constructed by standard genetic engineering techniques. By linking a soluble form of B7- 1, B7-2 or B7-3 to a toxin such as ricin, an agent capable of preventing T cell activation can be made. Infusion of one or a combination of immunotoxins, e.g., B7-2-ricin, 35 B7-1-ricin, into a patient may result in the death of T cells, particularly of activated T cells that express higher amounts of CD28 and CTLA4. Soluble forms of B7-2 in a monovalent WO 95/03408 : PCT/US94/08423 ~16~

form alone may be useful in blocking B7-2 function, as described above~ in which case a carrier molecule may also be employed.
Another method of preventing the function of a B Iymphocyte antigen is through the use of an antisense or triplex oligonucleotide. For example, an oligonucleotide S complement~ry to the area around the B7-1, B7-2 or B7-3 translation initiation site, (e.g., for B7-1, TGGCCCATGGCTTCAGA, (SEQ ID NO:20) nucleotides 326-309 and for B7-2, GCCAAAATGGATCCCCA (SEQ ID NO:21)), can be synthesized. One or more antisense oligonucleotides can be added to cell media, typically at 200 llg/ml, or ~Amini~tered to a patient to prevent the synthesis of B7-1, B7-2 and/or B7-3. The antisense oligonucleotide is 10 taken up by cells and hybridizes to the appro~liate B lymphocyte antigen mRNA to prevent translation. Alternatively, an oligonucleotide which binds double-stranded DNA to form a triplex construct to prevent DNA unwinding and transcription can be used. As a result of either, synthesis of one or more B lymphocyte antigens is blocked.

15 F Therapeutic Uses by Dow~re~ulation of Immune Respo~es Given the structure and function of the novel B lymphocyte antigens disclosed herein, it is possible to downregulate the function of a B lymphocyte antigen, and thereby downregulate immllne responses, in a number of ways. Downregulation may be in the form of inhibiting or blocking an immllne response already in progress or may involve preventing 20 the induction of an immllne response. The functions of activated T cells may be inhibited by ~u~ S~ g T cell responses or by inducing specific tolerance in T cells, or both.Immunosuppression of T cell responses is generally an active, non-antigen-specific, process which requires continuous exposure of the T cells to the s~l~s~ive agent. Tolerance, which involves inducing non-responsiveness or anergy in T cells, is distinguishable from 25 immunosuppression in that it is generally antigen-specific and persists after exposure to the tolerizing agent has ceased. Operationally, tolerance can be demonstrated by the lack of a T
cell response upon reexposure to specific antigen in the absence of the tolerizing agent.
Downregulating or preventing one or more B lymphocyte antigen functions, e.g., preventing high level lymphokine synthesis by activated T cells, will be useful in situations 30 of tissue, skin and organ transplantation and in graft-versus-host disease (GVHD). For example, blockage of T cell function should result in reduced tissue destruction in tissue transplantation. Typically, in tissue transplants, rejection of the kansplant is initiated through its recognition as foreign by T cells, followed by an immlme reaction that destroys the kansplant. The ~lmini~kation of a molecule which inhibits or blocks interaction of a B7 35 lymphocyte antigen with its natural ligand(s) on immllne cells (such as a soluble, monomeric form of a peptide having B7-2 activity alone or in conjunction with a monomeric form of a peptide having an activity of another B lymphocyte antigen (e.g., B7-1, B7-3) or blocking WO 95/03408 2 1 ~ 7 ~ ~ 1 PCT/US94/08423 antibody), prior to transplantation can lead to the binding of the molecule to the natural ligand(s) on the immune cells without tr~n~mitting the corresponding costimulatory signal.
Blocking B Iymphocyte antigen function in this manner prevents cytokine synthesis by imml-ne cells, such as T cells~ and thus acts as an immunosuppressant. Moreover, the lack of 5 costimulation may also be sufficient to anergize the T cells, thereby inducing tolerance in a subject. Induction of long-term tolerance by B Iymphocyte antigen-blocking reagents may avoid the necessity of repeated ~flmini~tration of these blocking reagents. To acheive sufficient immunosuppression or tolerance in a subject, it may also be necessary to block the function of a combination of B lymphocyte antigens. For example, it may be desirable to block the function of B7-2 and B7-1, B7-2 and B7-3, B7-1 and B7-3 or B7-2, B7-1 and B7-3 by ~-lmini~tering a soluble form of a combination of peptides having an activity of each of these antigens or a blocking antibody (separately or together in a single composition) prior to transplantation. Alternatively, inhibitory forms of B lymphocyte antigens can be used with other suppressive agents such as blocking antibodies against other T cell markers or against cytokines, other fusion proteins, e.g., CTLA41g, or immunosuppressive drugs.
The efficacy of particular blocking reagents in preventing organ transplant rejection or GVHD can be ~esesse~l using animal models that are predictive of efficacy in hllm~n~. The functionally important aspects of B7-1 are conserved structurally between species and it is therefore likely that other B lymphocyte antigens can function across species, thereby allowing use of reagents composed of human proteins in animal systems. Examples of ~I)rop.;ate systems which can be used include allogeneic cardiac grafts in rats and xenogeneic pancreatic islet cell grafts in mice, both of which have been used to examine the immunosuppressive effects of CTLA4Ig fusion proteins in vivo as described in Lenschow et al., Science, 257: 789-792 (1992) and Turka et al., Proc. Natl. Acad. Sci. USA, 89: 11102-11105 (1992). In addition, murine models of GVHD (see Paul ed., Fundamental Immunology, Raven Press, New York, 1989, pp. 846-847) can be used to determine the effect of blocking B Iymphocyte antigen function in vivo on the development of that disease.
Blocking B Iymphocyte antigen function, e.g., by use of a peptide having B7-2 activity alone or in combination with a peptide having B7-1 activity and/or a peptide having B7-3 activity, may also be therapeutically useful for treating autoimmune diseases. Many autoimmune disorders are the result of inapl)lol~liate activation of T cells that are reactive against self tissue and which promote the production of cytokines and autoantibodies involved in the pathology of the diseases. Preventing the activation of autoreactive T cells may reduce or elimin~te disease symptoms. Administration of reagents which blockcostimulation of T cells by disrupting receptor:ligand interactions of B Iymphocyte antigens can be used to inhibit T cell activation and prevent production of autoantibodies or T cell-derived cytokines which may be involved in the disease process. Additionally, blocking -wo 95/03408 ~ 91 - PCT/US94/08423 reagents may induce antigen-specific tolerance of autoreactive T cells which could lead to long-term relief from the disease. The efficacy of blocking reagents in preventing or alleviating autoimml-ne disorders can be determined using a number of well-characterized animal models of human autoimmnne ~ e~ees Examples include murine ~ue~ .ental S autoimmune encephalitis, systemic lupus erythmatosis in MRl llpr/lpr mice or NZB hybrid mice, murine autoimmune collagen arthritis, diabetes mellitus in NOD mice and BB rats, and murine experimental myaetheni~ gravis (see Paul ed., Fundamental Immunology, Raven Press, New York, 1989, pp. 840-856).
The IgE antibody response in atopic allergy is highly T cell dependent and, thus, inhibition of B lymphocyte antigen in~ e-1 T cell activation may be useful theld~tulically in the tre~tment of allergy and allergic reactions. An inhibitory form of B7-2 protein, such as a peptide having B7-2 activity alone or in combination with a peptide having the activity of another B lymphocyte antigen, such as B7-1, can be ~mini~tered to an allergic subject to inhibit T cell me~ te-l allergic responses in the subject. Inhibition of B lymphocyte antigen costim~llation of T cells may be accompagnied by exposure to allergen in conjunction with a~rop,iate MHC molecules.
Allergic reactions may be systemic or local in nature, depending on the route of entry of the allergen and the pattern of deposition of IgE on mast cells or basophils. Thus, it may be necessary to inhibit T cell me~liattod allergic responses locally or systemically by proper ~lmini~tration of an inhibitory form of B7-2 protein.
Inhibition of T cell activation through blockage of B lymphocyte antigen function may also be important therapeutically in viral infections of T cells. For example, in the acquired immune deficiency syndrome (AIDS), viral replication is stim~ tçA by T cell activation. Blocking B7-2 function could lead to a lower level of viral replication and thereby ameliorate the course of AIDS. In addition, it may also be nt~cess~ry to block the function of a combination of B lymphocyte antigens i.e., B7-1, B7-2 and B7-3. Surprisingly, HTLV-I infected T cells express B7-1 and B7-2. This expression may be important in the growth of HTLV-I infected T cells and the blockage of B7- 1 function together with the function of B7-2 and/or B7-3 may slow the growth of HTLV-I inc~l~cec~ lellk~mi~
Alternatively, stimlll~tion of viral replication by T cell activation may be in~ ced by contact with a stimnl~tQry form of B7-2 protein, for such purposes as generating retroviruses (e.g., various HIV isolates) in sufficient quantities for isolatation and use.

F. Therapeutic Uses by Upre~ulation of Tmmllne ~esponses Upregulation of a B lymphocyte antigen function, as a means of upregulating immune responses, may also be useful in therapy. Upregulation of immllne responses may be in the form of enhancing an existing immune response or eliciting an initial immune response. For WOg5/03~8 ~16 ~ O 91 PCT~S94/08423 example, enhancing an immune response through stimul~ting B lymphocyte antigen function may be useful in cases of viral infection. Viral infections are cleared primarily by cytolytic T
cells. In accordance with the present invention, it is believed that B7-2 and thus, B7-1 and B7-3 with their natural ligand(s) on T cells may result in an increase in the cytolytic activity S of at least some T cells. It is also believed that B7-2,B7-1, and B7-3 are involved in the initial activation and generation of CD8~ cytotoxic T cells. The addition of a soluble peptide having B7-2 activity, alone, or in combination with a peptide having the activity of another B
lymphocyte antigen, in a multi-valent form, to stim~ te T cell activity through the costimulation pathway would thus be therapeutically useful in situations where more rapid or 10 thorough clearance of virus would be beneficial. These would include viral skin diseases such as Herpes or shingles, in which cases the multi-valent soluble peptide having B7-2 activity or combination of such peptide and/or a peptide having B7-1 activity and/or a peptide having B7-3 activity is delivered topically to the skin. In addition, systemic viral diseases such as influenza, the common cold, and encephalitis might be alleviated by the 15 ~lmini.etration of stiml]l~tQry forms of B lymphocyte antigens systemically.
Altern~tively, anti-viral imml~ne responses may be enhanced in an infected patient by removing T cells from the patient, costimulating the T cells in vitro with viral antigen-pulsed APCs either ex~lc;ssillg a peptide having B7-2 activity (alone or in combination with a peptide having B7-1 activity and/or a peptide having B7-3 activity) or together with a 20 stim~ tory form of a soluble peptide having B7-2 activity (alone or in combination with a peptide having B7-1 activity and/or a peptide having B7-3 activity) and reintroducing the in vitro activated T cells into the patient. Another method of enhancing anti-viral immune responses would be to isolate infected cells from a patient, transfect them with a nucleic acid encoding a peptide having the activity of a B Iymphocyte antigen as described herein such that the cells express all or a portion of a B lymphocyte antigen on their surface, e.g., B7-2 or B7-3, and reintroduce the transfected cells into the patient. The infected cells would now be capable of delivering a costimlll~tory signal to, and thereby activate, T cells in vivo.
Stim~ tory forms of B lymphocyte antigens may also be used prophylactically in vaccines against various pathogens. Immunity against a pathogen, e.g., a virus, could be induced by vaccinating with a viral protein along with a stimulatory form of a peptide having B7-2 activity or another peptide having the activity of B lymphocyte antigen in an a~lupliate adjuvant. Alternately, an expression vector which encodes genes for both a pathogenic antigen and a peptide having the activity of a B lymphocyte antigen, e.g., a vaccinia virus expression vector engineered to express a nucleic acid encoding a viral protein and a nucleic acid encoding a peptide having B7-2 activity as described herein, can be used for vaccination. Present~tion of B7-2 with class I MHC proteins by, for example, a cell transfected to coexpress a peptide having B7-2 activity and MHC class I a chain protein and æ~7~9~ ~

~2 microglobulin may also result in activation of cytolytic CD8+ T cells and provide immunity from viral infection. Pathogens for which vaccines may be useful include hepatitis B, hepatitis C, Epstein-Barr virus, cytomegalovirus, HIV-1, HIV-2, tuberculosis, malaria and schistosomiasis.
In another aspect, a stimulatory form of one or more soluble peptides having an activity of a B lymphocyte antigen can be ~tlmini~tered to a tumor-bearing patient to provide a costim~ tQry signal to T cells in order to induce anti-tumor immllnity.

G. Modification of a Tumor Cell to Fxpress a Costimulatory Molecule The inability of a tumor cell to trigger a costimulatory signal in T cells may be due to a lack of e~s~,ei,~ion of a costiml-l~tory molecule, failure to express a costim~ tory molecule even though the tumor cell is capable of t;x~r~s~ lg such a molecule, insufficient expression of a costimulatory molecule on the tumor cell surface or lack of ex~r~s~ion of an appropriate costiml-l~t--ry molecule (e.g. ex~,les~ion of B7 but not B7-2 and/or B7-3). Thus, according to one aspect of the invention, a tumor cell is modified to express B7-2 and/or B7-3 by transfection of the tumor cell with a nucleic acid encoding B7-2 and/or B7-3 in a form suitable for ~,es~ion of B7-2 and/or B7-3 on the tumor cell surface. Alternatively, the tumor cell is modified by contact with an agent which induces or increases expression of B7-2 and/or B7-3 on the tumor cell surface. In yet another embodiment, B7-2 and/or B7-3is coupled to the surface of the tumor cell to produce a modified tumor cell. These and other emodiments are described in further detail in the following subsections.

(1). Tr~n~fection of a Tumor Cell with a Nucleic Acid Fn~oding a Costimulatory Molecule Tumor cells can be modified ex vivo to express B7-2 or B7-3, alone or in combination or in combination with B7-1 by transfection of isolated tumor cells with a nucleic acid encoding B7-2 and/or B7-3 and B7-1 in a form suitable for ~ s~.ion of the molecule on the surface of the tumor cell. The terms "transfection" or "transfected with" refers to the introduction of exogenous nucleic acid into a m~mm~ n cell and encompass a variety of techniques useful for introduction of nucleic acids into m~nnm~ n cells including electroporation, calciurn-phosphate precipitation, DEAE-dextran treatment, lipofection, microinjection and infection with viral vectors. Suitable methods for transfecting m~mm~ n cells can be found in Sarnbrook et al. (Molec~ r Clonin~: A T ~horatory ~anuaL
~nd F.rlition, Cold Spring Harbor Laboratory press (1989)) and other laboratory textbooks.
The nucleic acid to be introduced may be, for example, DNA encompassing the gene(s) encoding B7-2 and/or B7-3, sense strand RNA encoding B7-2 and/or B7-3 or a recombinant WO 95lO3408 21 6 7 0 91 pcTluss4los423 expression vector containing a cDNA encoding B7-2 and/or B7-3. The nucleotide sequence of a cDNA encoding human B7-2 is shown in the Sequence Listing.
A plefe,l~d approach for introducing nucleic acid encoding B7-2 and/or B7-3 intotumor cells is by use of a viral vector cont~ining nucleic acid, e.g. a cDNA, encoding B7-2 and/or B7-3. Examples of viral vectors which can be used include retroviral vectors (Eglitis, M.A., et al., Science 230, 1395-1398 (1985); Danos, O. and Mulligan, R., Proc. Natl. Acad.
Sci. USA 85, 6460-6464 (1988); Markowitz, D., et al., J. Virol. 62, 1120-1124 (1988)), adenoviral vectors (Rosenfeld, M.A., et al., Cell 68, 143-155 (1992)) and adeno-associated viral vectors (Tratschin, J.D., et al., Mol. Cell. Biol. 5, 3251-3260 (1985)). Infection of tumor cells with a viral vector has the advantage that a large proportion of cells will receive nucleic acid, thereby obviating a need for selection of cells which have received nucleic acid, and molecules encoded within the viral vector, e.g. by a cDNA contained in the viral vector, are expressed efficiently in cells which have taken up viral vector nucleic acid.
Alternatively, B7-2 and/or B7-3 can be expressed on a tumor cell using a plasmidexl ,es~ion vector which contains nucleic acid, e.g. a cDNA, encoding B7-2 and/or B7-3.
Suitable plasmid ~x~ ssion vectors include CDM8 (Seed, B., Nature 329, 840 (1987)) and pMT2PC (~ n, et al., EMBO J. 6, 187- 195 (1987)). Suitable vectors and methods for ~x~le3~ing nucleic acids in host cells, such as tumor cells are described in further detail herein.
When transfection of tumor cells leads to modification of a large proportion of the tumor cells and efficient expression of B7-2 and/or B7-3 on the surface of tumor cells, e.g.
when using a viral ex~leSSiOn vector, tumor cells may be used without further isolation or subcloning. Alternatively, a homogenous population of transfected tumor cells can be prepared by isolating a single transfected tumor cell by limitin~ dilution cloning followed by expansion of the single tumor cell into a clonal population of cells by standard techniques.

(2). Tn~ tion or Jncr~ed F~ression of a Costimulatory Molecule on a T-lmor Cell Surface A tumor cell can be modified to trigger a costim--l~tory signal in T cells by inducing or increasing the level of expression of B7-2 and/or B7-3 on a tumor cell which is capable of expressing B7-2 and/or B7-3 but fails to do so or which expresses insufficient amounts of B7-2 and/or B7-3 to activate T cells. An agent which stimulates expression of B7-2 and/or B7-3 can be used in order to induce or increase expression of B7-2 and/or B7-3 on the tumor cell surface. ~or example, tumor cells can be contacted with the agent in vitro in a culture medium. The agent which stimulates expression of B7-2 and/or B7-3 may act, for instance, by increasing transcription of B7-2 and/or B7-3 gene, by increasing translation of B7-2 and/or B7-3 mRNA or by increasing stability or transport of B7-2 and/or B7-3 to the cell WO 95t03408 ~16 ~ ~ ~1 PCT/US94/08423 s-lrf~ce For example, it is known that expression of B7 can be upregulated in a cell by a second messenger pathway involving cAMP. Nabavi, N., et al. Nature 360, 266-268 (1992).
B7-2 and B7-3 may likewise be inducible by cAMP. Thus, a tumor cell can be contacted with an agent, which increases intracellular cAMP levels or which mimics cAMP, such as a 5 cAMP analogue, e.g. dibutyryl cAMP, to stim~ te expression of B7-2 and/or B7-3 on the tumor cell surface. It is also known that expression of B7 can be in(l~lcecl on normal resting B
cells by cro.s~linking cell-surface MHC class II molecules on the B cells with an antibody against the MHC class II molecules. Kuolova, L., et al., J. Exp. Med 173, 759-762 (1991).
Similarly, B7-2 and B7-3 can be in~ ecl on resting B cells by crosslinking cell-surface MHC
10 class II molecules on the B cells. Accordingly, a tumor cell which expresses MHC class Il molecules on its surface can be treated with anti-MHC class II antibodies to induce or increase B7-2 and or B7-3 ex~les~ion on the tumor cell surface. In addition, interleukin-4 (IL-4) which has been found to induce expression of B7-2 on B cells, may be used to upregulate expression of B7-2 on tumor cells (Stack R.M., et al., J. Cell. Biochem. Suppl 1(18):434 (1994).
Another agent which can be used to induce or increase expression of B7-2 and/or B7-3 on a tumor cell surface is a nucleic acid encoding a transcription factor which upregulates transcription of the gene encoding the costimulatory molecule. This nucleic acid can be transfected into the tumor cell to cause increased transcription of the costim~ tory molecule 20 gene, resulting in increased cell-surface levels of the costimulatory molecule.

(3). Couplir~ of a Costimulatory Molecule to the Sllrface of a Tllmor Cell In another embodiment, a tumor cell is modified to be capable of triggering a costim~ tory signal in T cells by coupling B7-2 and/or B7-3 to the surface of the tumor cell.
25 For example, B7-2 and/or B7-3 molecules can be obtained using standard recombinant DNA
technology and ~ s~ion systems which allow for production and isolation of the costim~ tory molecule(s). Altern~tively, B7-2 and/or B7-3 can be isolated from cells which express the costimlll~tory molecule(s) using standard protein purification techniques. For example, B7-3 protein can be isolated from activated B cells by immunoprecipitation with an 30 anti-B7-3 antibody such as the BB1 monoclonal antibody. The isolated costimlll~tory molecule is then coupled to the tumor cell. The terms "coupled" or "coupling" refer to a chemical, enzymatic or other means (e.g., antibody) by which B7-2 and/or B7-3 is linked to a tumor cell such that the costimulatory molecule is present on the surface of the tumor cell and is capable of triggering a costimulatory signal in T cells. For example, B7-2 and/or B7-3 can 35 be chemically crosslinked to the tumor cell surface using commercially available cro~slinking reagents (Pierce, Rockford IL). Another approach to coupling B7-2 and/or B7-3 to a tumor cell is to use a bispecific antibody which binds both the costim~ tory molecule and a cell-WO 95/0340~ 2 1 6 7 0 9 ~ PCT/US94/08423 surface molecule on the tumor cell. Fragments, mutants or variants of B7-2 and/or B7-3 which retain the ability to trigger a costim~ tQry signal in T cells when coupled to the surface of a tumor cell can also be used.

(4). Mo~lificatio~ of T-lmor Cell~ to Fxpress Multiple Costimulatory Molecules Another aspect of the invention is a tumor cell modified to express multiple costimlll~tory molecules. The temporal ~ s~ion of costim~ tory molecules on activated B cells is different for B7, B7-2 and B7-3. For example, B7-2 is expressed early following B
cell activation, whereas B7-3 is ~ essed later. The different costim~ tory molecules may thus serve distinct functions during the course of an immllne response. An effective T cell response may require that the T cell receive costim~ tQry signals from multiple costimulatory molecules. Accordingly, the invention encompasses a tumor cell which is modified to express more than one costimlll~tQry molecule. For example, a tumor cell can be modified to express both B7-2 and B7-3. ~It~rn~tively, a tumor cell modified to express B7-2 can be further modified to express B7-1. Similarly, a tumor cell modified to express B7-3 can be further modified to express B7-1. A tumor cell can also be modified to express B7-1, B7-2 and B7-3. A tumor cell can be modified to express multiple costimulatory molecules (e.g., B7- 1 and B7-2) by any of the techniques described herein.
Before modification, a tumor cell may not express any costimulatory molecules, or may express certain costimulatory molecules but not others. As described herein, tumor cells can be modified by transfecting the tumor cell with nucleic acid encoding a costimulatory molecule(s), by inducing the c;~s~les~ion of a costim~ tory molecule(s) or by coupling a costimulatory molecule(s) to the tumor cell. For example, a tumor cell transfected with nucleic acid encoding B7-2 can be further transfected with nucleic acid encoding B7-1. The cDNA sequence and d~ ce~1 amino acid sequence of human B7-1 is shown in the Sequence T .i~ting Alten ~tively, more than one type of modification can be used. For example, a tumor cell transfected with a nucleic acid encoding B7-2 can be stimulated with an agent which in~ ces t;~ s~ion of B7-1.

(5) Additio~l Motlifiçation of a Tllmor Cell to F~ress MHC Molec--les Another aspect of this invention features modified tumor cells which express a costimulatory molecule and which express one or more MHC molecules on their surface to trigger both a costimulatory signal and a primary, antigen-specific, signal in T cells. Before modification, tumor cells may be unable to express MHC molecules, may fail to express MHC molecules although they are capable of e~les~ing such molecules, or may express insufficient amounts of MHC molecules on the tumor cell surface to cause T cell activation.
Tumor cells can be modified to express either MHC class I or MHC class II molecules, or 2~09i ~

both. One approach to modifying tumor cells to express MHC molecules is to transfect the tumor cell with one or more nucleic acids encoding one or more MHC molecules.
Alternatively, an agent which induces or increases expression of one or more MHCmolecules on tumor cells can be used to modify tumor cells. Inducing or increasing 5 ~ s~ion of MHC class II molecules on a tumor cell can be particularly beneficial for activating CD4+ T cells against the tumor since the ability of MHC class II+ tumor cells to directly present tumor peptides to CD4+ T cells bypasses the need for professional MHC
class II+ APCs. This can improve tumor immunogenicity because soluble tumor antigen (in the form of tumor cell debris or secreted protein) may not be available for uptake by 10 professional MHC class II + APCs.
One embodiment of the invention is a modified tumor cell which expresses B7-2 and/or B7-3 and one or more MHC class II molecules on their cell surface. MHC class II
molecules are cell-surface al~ heterodimers which structurally contain a cleft into which antigenic peptides bind and which function to present bound peptides to the antigen-specific 15 TcR. Multiple, different MHC class II proteins are e~les~ed on professional APCs and different MHC class II proteins bind different antigenic peptides. Expression of multiple MHC class II molecules, therefore, increases the spectrum of antigenic peptides that can be presented by an APC or by a modified tumor cell. The a and ,B chains of MHC class II
molecules are encoded by dirrt l~llL genes. For instance, the hurnan MHC class II protein 20 HLA-DR is encoded by the HLA-DRa and HLA-DR,~ genes. Additionally, many polymorphic alleles of MHC class II genes exist in human and other species. T cells of a particular individual respond to stimulation by antigenic peptides in conjunction with self MHC molecules, a phenomenon termed MHC restriction. In addition, certain T cells can also respond to stim~ tion by polymorphic alleles of MHC molecules found on the cells of other 25 individuals, a phenomenon termed allogenicity. For a review of MHC class II structure and function, see Germain and Margulies, Ann. Rev. Immunol. 1 1: 403-450, 1993.
Another embodiment of the invention is a modified tumor cell which expresses B7-2 and/or B7-3 and one or more MHC class I molecules on the cell surface. Similar to MHC
class II genes, there are multiple MHC class I genes and many polymorphic alleles of these 30 genes are found in human and other species. Like MHC class II proteins, class I proteins bind peptide fr~gment~ of antigens for presentation to T cells. A functional cell-surface class I molecule is composed of an MHC class I a chain protein associated with a ~2-microglobulin protein.

(6). Tran~fection of a Tumor Cell with Nucleic Acid F.nco~lin~ M~IC Molecules Tumor cells can be modified ex vivo to express one or more MHC class II molecules by transfection of isolated tumor cells with one or more nucleic acids encoding one or more ~WO 95/03408 21 6 ~ O ~1 PCT/US94/08423 MHC class II a chains and one or more MHC class II ~ chains in a form suitable for ~x~ ssion of the MHC class II molecules(s) on the surface of the tumor cell. Both an a and a ,B chain protein must be present in the tumor cell to form a surface heterodimer and neither chain will be expressed on the cell surface alone. The nucleic acid sequences of many murine and human class II genes are known. For examples see Hood, L., et al. Ann. Rev. Immunol. 1, 529-568 (1983) and Auffray, C. and Strominger, J.L., Advances in Human Genetics 15, 197-247 (1987). Preferably, the introduced MHC class II molecule is a selfMHC class II
molecule. Alternatively, the MHC class II molecule could be a foreign, allogeneic, MHC
class II molecule. A particular foreign MHC class II molecule to be introduced into tumor cells can be selected by its ability to induce T cells from a tumor-bearing subject to proliferate and/or secrete cytokines when stimulated by cells expressing the foreign MHC
class II molecule (i.e. by its ability to induce an allogeneic response). The tumor cells to be transfected may not express MHC class II molecules on their surface prior to transfection or may express amounts insufficient to stim~ t~ a T cell response. Alternatively, tumor cells which express MHC class II molecules prior to transfection can be further transfected with additional, different MHC class II genes or with other polymorphic alleles of MHC class II
genes to increase the spectrum of antigenic fr~gment~ that the tumor cells can present to T
cells.
Fr~ment.~, mutants or variants of MHC class II molecules that retain the ability to bind peptide antigens and activate T cell responses, as evidenced by proliferation and/or lymphokine production by T cells, are considered within the scope of the invention. A
preferred variant is an MHC class II molecule in which the cytoplasmic domain of either one or both of the a and ~ chains is tr~lnc~t~A It is known that truncation of the cytoplasmic domains allows peptide binding by and cell surface ex~le3~ion of MHC class II molecules but prevents the induction of endogenous B7 ~x~les~ion, which is triggered by an intracellular signal generated by the cytoplasmic domains of the MHC class II protein chains upon crosslinking of cell surface MHC class II molecules. Kuolova. L., et al., J. Exp. Med. 173, 759-762 (1991), Nabavi, N., et al. Nature 360, 266-268 (1992). Expression of B7-2 and B7-3 is also in~ recl by crosslinking surface MHC class II molecules, and thus truncation of MHC
class II molecules may also prevent induction of B7-2 and/or B7-3. In tumor cells transfected to constitutively express B7-2 and/or B7-3, it may be desirable to inhibit the expression of - endogenous costimulatory molecules, for instance to restrain potential downregulatory fee~lb~ck mech~ni~m~ Transfection of a tumor cell with a nucleic acid(s) encoding a cytoplasmic domain-trllnr~tt~l form of MHC class II a and ,B chain proteins would inhibit endogenous B7-1 expression and possibly also endogenous B7-2 and B7-3 expression. Such variants can be produced by, for example, introducing a stop codon in the MHC class II chain gene(s) after the nucleotides encoding the transmembrane spanning region. The cytoplasmic 2~0~1 ~

domain of either the a chain or the ,B chain protein can be truncated, or~ for more complete inhibition of B7 (and possibly B7-2 and/or B7-3) induction, both the a and ,B chains can be kuncated. See e.g. Griffith et al., Proc. Natl. Acad. Sci US~l 85: 4847-4852, (1988), Nabavi et al., J. Immunol. 142: 1444-1447, (1989).
S Turnor cells can be modified to express an MHC class I molecule by kansfection with a nucleic acid encoding an MHC class I a chain protein. For examples of nucleic acids see Hood, L., et al. Ann. Rev. ImmunoL 1, 529-568 (1983) and Auffray, C. and Strominger, J.L., Advances in Human Genetics 15, 197-247 (1987). Optionally, if the tumor cell does not express ,~-2 microglobulin, it can also be kansfected with a nucleic acid encoding the ~-2 microglobulin protein. For examples of nucleic acids see Gussow, D., et al., J. Immunol. 139, 3132-3138 (1987) and Parnes, J.R., et al., Proc. Natl. Acad. Sci. USA 78, 2253-2257 (1981).
As for MHC class II molecules, increasing the number of different MHC class I genes or polymorphic alleles of MHC class I genes expressed in a tumor cell can increase the spectrum of antigenic fr~gment~ that the turnor cells can present to T cells.
When a tumor cell is kansfected with nucleic acid which encodes more than one molecule, for example a B7-2 and/or B7-3 molecule(s), an MHC class II a chain protein and an MHC class II ,B chain protein, the transfections can be performed simultaneously or sequentially. If the transfections are performed ~iml-lt~neously, the molecules can be introduced on the same nucleic acid, so long as the encoded sequences do not exceed a carrying capacity for a particular vector used. Alternatively, the molecules can be encoded by separate nucleic acids. If the kansfections are con~ cte~l sequentially and tumor cells are selected using a selectable marker, one selectable marker can be used in conjunction with the first inkoduced nucleic acid while a dirr~ l selectable marker can be used in conjunction with the next introduced nucleic acid.
The expression of MHC molecules (class I or class II) on the cell surface of a turnor cell can be determined, for example, by immllnoflourescence of tumor cells usingfluorescently labeled monoclonal antibodies directed against different MHC molecules.
Monoclonal antibodies which recognize either non-polymorphic regions of a particular MHC
molecule (non-allele specific) or polymorphic regions of a particular MHC molecule (allele-specific) can be used and are known to those skilled in the art.

(7). In~ tion or Tncreased Fxpression of MHC Molecules on a Tumor Cell Another approach to modifying a tumor cell ex vivo to express MHC molecules on the surface of a tumor cell is to use an agent which stimulates expression of MHC molecules in order to induce or increase expression of MHC molecules on the tumor cell surface. For example, tumor cells can be contacted with the agent in vitro in a culture medium. An agent which stimul~tes expression of MHC molecules may act, for instance, by increasing

8 PCT/US94/08423 2l~a~l transcription of MHC class I and/or class II genes, by increasing translation of MHC class I
and/or class II mRNAs or by increasing stability or transport of MHC class I and/or class Il proteins to the cell surface. A nurnber of agents have been shown to increase the level of cell-surface expression of MHC class II molecules. See for example Cockfield, S.M. et al., J.
5 Immunol. 144, 2967-2974 (1990); Noelle, R.J. et al. J. ImmunoL 137, 1718-1723 (1986);
Mond, J.J., et al., J. lmmunol. 127, 881-888 (1981); Willman, C.L., et al. J. Exp. Med., 170, 1559-1567 (1989); Celada, A.and Maki, R. J. Immunol. 146, 114-120 (1991) and Glimcher, L.H. and Kara, C.J. Ann. Rev. Immunol. 10, 13-49 (1992) and references therein. These agents include cytokines, antibodies to other cell surface molecules and phorbol esters. One agent which upregulates MHC class I and class II molecules on a wide variety of cell types is the cytokine interferon-~. Thus, for example, tumor cells modif1ed to express B7-2 and/or B7-3 and B7- 1 can be further modified to increase ~ cssion of MHC molecules by contact with interferon-~.
Another agent which can be used to induce or increase ex~l~ssion of an MHC
molecule on a tumor cell surface is a nucleic acid encoding a transcription factor which upregulates transcription of MHC class I or class II genes. Such a nucleic acid can be transfected into the tumor cell to cause increased transcription of MHC genes, resulting in increased cell-surface levels of MHC proteins. MHC class I and class II genes are regulated by different transcription factors. However, the multiple MHC class I genes are regulated coordinately, as are the multiple MHC class II genes. Therefore, transfection of a tumor cell with a nucleic acid encoding a transcription factor which regulates MHC gene expression may increase e~res~ion of several different MHC molecules on the tumor cell surface.
Several transcription factors which regulate the expression of MHC genes have been identified, cloned and characterized. For example, see Reith, W. et al., Genes Dev. 4, 1528-1540, (1990); Liou, H.-C., et al., Science 247, 1581-1584 (1988), Didier, D.K., et al., Proc.
Natl. Acad. Sci. US~ 85, 7322-7326 (1988).

(8). Inhibition of Invari~nt Ch~in Fxpression in Tllmor Cell~
Another embodiment of the invention provides a tumor cell modified to express a T
cell costimulatory molecule (e.g., B7-2 and/or B7-3 and B7-1) and in which expression of an MHC class II-associated protein, the invariant chain, is inhibited. Invariant chain expression is inhibited to promote association of endogenously-derived TAA peptides with MHC class II
molecules to create an antigen-MHC complex. This complex can trigger an antigen-specific signal in T cells to induce activation of T cells in conjunction with a costimulatory signal.
MHC class II molecules have been shown to be capable of pres~ontinP endogenously-derived peptides. Nuchtern, J.G., et al. Nature 343, 74-76 (1990); Weiss, S. and Bogen, B. Cell 767-776 (1991). However, in cells which naturally express MHC class II molecules, the a and ,B

WO 95/03408 ;,~ Q 9 l PCT/US94/08423 chain proteins are associated with the invariant chain (hereafter Ii) during intracellular transport of the proteins from the endoplasmic reticulum. It is believed that Ii functions in part by preventing the association of endogenously-derived peptides with MHC class II
molecules. Elliott, W., et al. J. Immunol. 138, 2949-2952 (1987); Stockinger, B., et al. Cell 56, 683-689 (1989); Guagliardi, L., et al. Nature (London) 343, 133-139 (1990); Bakke, O., et al. Cell 63, 707-716 (1990); Lottreau, V., et al. Nature 348,600-605 (1990); Peters, J., et al. Nature 349, 669-676 (1991); Roche, P., et al.Nature 345, 615-618 (1990); Teyton, L., et al. Nature 348, 39-44 (1990). Since TAAs are synthesized endogenously in tumor cells, peptides derived from them are likely to be available intracellularly. Accordingly, inhibiting the ex~,~s~ion of Ii in tumor cells which express Ii may increase the likelihood that TAA
peptides will associate with MHC class II molecules. Consistent with this mech~ni~m, it was shown that supertransfection of an MHC class II+, Ii- tumor cell with the Ii gene prevented stim~ tion of tumor-specific immllnity by the tumor cell. Clements, V.K., et al. J. Immunol.
149, 2391-2396 (1992).
Prior to modification, the ~ rt;s~ion of Ii in a tumor cell can be assessed by detecting the presence or absence of Ii mRNA by Northern blotting or by detecting the presence or ~bsence of Ii protein by imml-noprecipitation. A preferred approach for inhibiting ex~ ssion of Ii is by introducing into the tumor cells a nucleic acid which is antisense to a coding or regulatory region of the Ii gene, which have been previously described. Koch, N., et al., EMBO J. 6, 1677-1683, (1987). For example7 an oligonucleotide complement~ry to nucleotides near the translation initiation site of the Ii mRNA can be synth~si7P-l One or more antisense oligonucleotides can be added to media cont~inin~ tumor cells, typically at a concentration of oligonucleotides of 200 ,~Lg/ml. The ~ntisçn~e oligonucleotide is taken up by tumor cells and hybridizes to Ii mRNA to prevent translation. In another embodiment, a recombinant expression vector is used in which a nucleic acid encoding sequences of the Ii gene in an orientation such that mRNA which is ~nti~çn~e to a coding or regulatory region of the Ii gene is produced. Tumor cells transfected with this recombinant expression vector thus contain a continuous source of Ii ~nfi~çn~e nucleic acid to prevent production of Ii protein.
~ltern~tively, Ii expression in a tumor cell can be inhibited by treating the tumor cell with an agent which interferes with Ii expression. For example, a ph~ ceutical agent which inhibits Ii gene c~ es~ion, Ii mRNA translation or Ii protein stability or intracellular transport can be used.

(9). Types of Tl-mor Cells to be Modified The tumor cells to be modified as described herein include tumor cells which can be transfected or treated by one or more of the approaches encompassed by the present invention to express B7-2 and/or B7-3, alone or in combination with B7-1. If necessary, the tumor wo 9~,03408 ~ 9 ~ PCT/US94108423 cells can be further modified to express MHC molecules or an inhibitor of Ii expression. A
tumor from which tumor cells are obtained can be one that has arisen spontaneously, e.g in a human subject, or may be experimentally derived or induced, e.g. in an animal subject. The tumor cells can be obtained, for example, from a solid tumor of an organ, such as a tumor of 5 the lung, liver, breast, colon, bone etc. ~lign~ncies of solid organs include carcinomas, sarcomas, melanomas and neuroblastomas. The tumor cells can also be obtained from a blood-borne (ie. dispersed) m~ n~ncy such as a lymphoma, a myeloma or a leukemia.
The tumor cells to be modified include those that express MHC molecules on theircell surface prior to transfection and those that express no or low levels of MHC class I
and/or class II molecules. A minority of normal cell types express MHC class II molecules.
It is therefore expected that many tumor cells will not express MHC class II molecules naturally. These tumors can be modified to express B7-2 and/or B7-3 and MHC class II
molecules. Several types of tumors have been found to naturally express surface MHC class II molecules, such as melanomas (van Duinen et al., Cancer Res. 48, 1019-1025, 1988), diffuse large cell lymphomas (O'Keane et al., Cancer 66, 1147-1153, 1990), squamous cell carcinomas of the head and neck (Mattijssen et al., Int. J. Cancer 6, 95- l O0, 1991) and colorectal carcinomas (Moller et al., Int. J. Cancer 6, 155-162, 1991). Tumor cells which naturally express class II molecules can be modified to express B7-2 and/or B7-3, and, in addition, other class II molecules which can increase the spectrum of TAA peptides which can be presented by the tumor cell. Most non-m~ n~nt cell types express MHC class I
molecules. However, m~lign~nt transformation is often accompanied by downregulation of expression of MHC class I molecules on the surface of tumor cells. Csiba, A., et al., Brit. J.
Cancer 50, 699-709 (1984). Importantly, loss of expression of MHC class I antigens by tumor cells is associated with a greater aggressiveness and/or metastatic potential of the tumor cells. Schrier, P.I., et al. Nature 305, 771-775 (1983); Holden, C.A., et al. J. Am. Acad.
Dermatol. 9., 867-871 (1983); Baniyash, M., et al. J Immunol. 129, 1318-1323 (1982).
Types of tumors in which MHC class I expression has been shown to be inhibited include melarlomas, colorectal carcinomas and squarnous cell carcinomas. van Duinen et al., Cancer Res. 48, 1019-1025, (1988); Moller et al., Int. J. Cancer 6, 155-162, (1991), Csiba, A., et al., Brit. J. Cancer 50, 699-709 (1984); Holden, C.A., et al. J. Am. Acad. Dermatol. 9., 867-871 (1983). A tumor cell which fails to express class I molecules or which expresses only low levels of MHC class I molecules can be modified by one or more of the techniques described herein to induce or increase expression of MHC class I molecules on the tumor cell surface to enhance tumor cell immlmngenicity.

WO 95/03408 PCTtUS94/08423 2~7~ 50

(10). Modification of Tllmor Cells In Vivo Another aspect of the invention provides methods for increasing the immunogenicity of a tumor cell by modification of the tumor cell in vivo to express B7-2 and/or B7-3 and B7- ~
1 to trigger a costimulatory signal in T cells. In addition, tumor cells can be further modified in vivo to express MHC molecules to trigger a primary, antigen-specific, signal in T cells.
Tumor cells can be modified in vivo by introducing a nucleic acid encoding B7-2 and/or B7-3 and B7-1 into the tumor cells in a form suitable for expression of the costimnl~t-)ry molecule(s) on the surface of the tumor cells. Likewise, nucleic acids encoding MHC class I
or class II molecules or an ~nti~Pn~e sequence of the Ii gene can be introduced into tumor cells in vivo. In one embodiment, a recombinant ~x~les~ion vector is used to deliver nucleic acid encoding B7-2 and/or B7-3 and B7-1 to tumor cells in vivo as a form of gene therapy.
Vectors useful for in vivo gene therapy have been previously described and include retroviral vectors, adenoviral vectors and adeno-associated viral vectors. See e.g. Rosenfeld, M.A., Cell 68, 143-155 (1992); Anderson, W.F., Science 226, 401-409 (1984); Friedman, T., Science 244, 1275-1281 (1989). Alternatively, nucleic acid can be delivered to tumor cells in vivo by direct injection of naked nucleic acid into tumor cells. See e.g. Acsadi, G., et al., Nafure 332, 815-818 (1991). A delivery ayy~lus is commercially available (BioRad).
Optionally, to be suitable for injection, the nucleic acid can be complexed with a carrier such as a liposome. Nucleic acid encoding an MHC class I molecule complexed with a liposome has been directly injected into tumors of melanoma patients. Hoffman, M., Science 256, 305-309 (1992)-Tumor cells can also be modified in vivo by use of an agent which induces or increases ~;x~,c;s~ion of B7-2 and/or B7-3 and B7-1 (and, if necessary, MHC molecules) as described herein. The agent may be ~lmini.~tered systemically, e.g. by inll~v~llous injection, or, preferably, locally to the tumor cells.

(11). The Fffector Ph~ of the Anti-TIlmor T Cell-Mediated ~mmune Response The modified tumor cells of the invention are useful for stimnl~ting an anti-tumor T
cell-mediated immune response by triggering an antigen-specific signal and a costimlll~tory signal in tumor-specific T cells. Following this inductive, or afferent, phase of an immlm~
response, effector populations of T cells are generated. These effector T cell populations can include both CD4+ T cells and CD8+ T cell. The effector populations are responsible for elimin~tion of tumors cell, by, for example, cytolysis of the tumor cells. Once T cells are activated, ~;xylt;;s~ion of a costim~ tory molecule is not required on a target cell for recognition of the target cell by effector T cells or for the effector functions of the T cells.
Harding, F.A. and Allison, J.P. J. ~xp. Med. 177, 1791-1796 (1993). Therefore, the anti-tumor T cell-mç~i~tecl immune response inclll~ecl by the modified tumor cells of the invention WO 95/03408 2 ~ 6 ~ ~ ~1 PCT/tJS94/08423 is effective against both the modified tumor cells and unrnodified tumor cells which do not express a costim~ tory molecule.
- Additionally, the density and/or type of MHC molecules on the cell surface required for the afferent and efferent phases of a T cell-mediated immune response can differ. Fewer S MHC molecules, or only certain types of MHC molecules (e.g. MHC class I but not MHC
class II) may be needed on a tumor cell for recognition by effector T cells than is needed for the initial activation of T cells. Therefore, tumor cells which naturally express low amounts of MHC molecules but are modified to express increased amounts of MHC molecules can induce a T cell-mediated imm~lne response which is effective against the unmodified tumor 10 cells. Alt~rn~tively, tumor cells which naturally express MHC class I molecules but not MHC class II molecules which are then modified to express MHC class II molecules can induce a T cell-mediated imm~lne response which includes effector T cell populations which can elimin~te the parental MHC class I+, class II- tumor cells.

(12). Therapeutic Co~ )osilions of Tllmor Cells Another aspect of the invention is a composition of modified tumor cells in a biologically compatible form suitable for ph~rm~ce-ltical ~-lminictration to a subject in vivo.
This composition compri~es an amount of modified tumor cells and a physiologically acceptable carrier. The amount of modified tumor cells is selected to be therapeutically 20 effective. The term "biologically compatible form suitable for ph~rm~ceutical ~tlmini~tration in vivo" means that any toxic effects of the tumor cells are outweighed by the therapeutic effects of the tumor cells. A "physiologically acceptable carrier" is one which is biologically compatible with the subject. Exarnples of acceptable carriers include saline and aqueous buffer solutions. In all cases, the compositions must be sterile and must be fluid to the extent 25 that easy syringability exists. The term "subject" is inten~le~l to include living org~ni~m~ in which tumors can arise or be experiment~lly in~ e~1 Examples of subjects include hllm~n~
dogs, cats, mice, rats, and transgenic species thereof.
A~lmini~tration of the therapeutic compositions of the present invention can be carried out using known procedures, at dosages and for periods of time effective to achieve the 30 desired result. For example, a therapeutically effective dose of modified tumor cells may vary according to such factors as age, sex and weight of the individual, the type of tumor cell and degree of tumor burden, and the immunological competency of the subject. Dosage regimens may be adjusted to provide optimum therapeutic responses. For instance, a single dose of modified tumor cells may be ~lmini~tt?red or several doses may be ~-~mini~tered over 35 time. Admini~tration may be by injection, including intravenous, intramuscular, a~e~;Loneal and subcutaneous injections.

W095/03~8 PCT~S94/08423 9~ ~

(13). Activation of Tllmor-specific T J,ymphocytes In Vitro Another approach to inducing or enhancing an anti-tumor T cell-mediated immune response by triggering a costimulatory signal in T cells is to obtain T lymphocytes from a tumor-bearing subject and activate them in vitro by stimulating them with tumor cells and a stim~ tory form of B7-2 and/or B7-3, alone or in combination with B7-1. T cells can be obtained from a subject, for example, from peripheral blood. Peripheral blood can be further fractionated to remove red blood cells and enrich for or isolate T lymophocytes or T
lymphocyte subpopulations. T cells can be activated in vitro by culturing the T cells with tumor cells obtained from the subject (e.g. from a biopsy or from peripheral blood in the case of blood-borne m~lign~ncies) together with a stimulatory form of B7-2 and/or B7-3 or, t~rn~tively, by exposure to a modified tumor cell as described herein. The term "stim~ tory form" means that the costimulatory molecule is capable of cro~linking its receptor on a T cell and triggering a costimtll~tory signal in T cells. The stim~ tory form of the costimulatory molecule can be, for example, a soluble multivalent molecule or an immobilized form of the costimulatory molecule, for instance coupled to a solid support.
Fr~gment~, mllt~nt~ or variants (e.g. fusion proteins) of B7-2 and/or B7-3 which retain the ability to trigger a costimlll~tory signal in T cells can also be used. In a plefellcd embodiment, a soluble extracellular portion of B7-2 and/or B7-3is used to provide costimlll~tion to the T cells. Following culturing of the T cells in vitro with tumor cells and B7-2 and/or B7-3, or a modified tumor cell, to activate tumor-specific T cells, the T cells can be ~lmini~t~red to the subject, for example by inkavenous injection.

(~4). Therapeutic Uses of Mo-1ified Tllmor Cells The modified tumor cells of the present invention can be used to increase tumor immunogenicity, and therefore can be used therapeutically for inducing or enhancing T
lymphocyte-me(li~te-l anti-tumor immunity in a subject with a tumor or at risk of developing a tumor. A method for keating a subject with a tumor involves obtaining tumor cells from the subject, modifying the tumor cells ex vivo to express a T cell costimlll~tory molecule, for example by transfecting them with an al,plu~,;ate nucleic acid, and ~lmini~tering a therapeutically effective dose of the modified tumor cells to the subject. Appropriate nucleic acids to be inkoduced into a tumor cell include nucleic acids encoding B7-2 and/or B7-3, alone or together with nucleic acids encoding B7-l,MHC molecules (class I or class II) or Ii antisense sequences as described herein. Alternatively, after tumor cells are obtained from a subject, they can be modified ex vivo using an agent which induces or increases expression of B7-2 and/or B7-3 (and possibly also using agent(s) which induce or increase B7-1 or MHC
molecules).

Tumor cells can be obtained from a subject by, for example, surgical removal of tumor cells, e.g. a biopsy of the tumor, or from a blood sample from the subject in cases of blood-borne m~lign~ncies. In the case of an experimentally in-luced tumor, the cells used to induce the tumor can be used, e.g. cells of a tumor cell line. Samples of solid tumors may be 5 treated prior to modification to produce a single-cell suspension of tumor cells for m~xim~l efficiency of transfection. Possible tre~tment~ include manual dispersion of cells or enzymat;c digestion of connective tissue fibers, e.g. by collagenase.
Tumor cells can be transfected immediately after being obtained from the subject or can be cultured in vitro prior to transfection to allow for further ch~r~clel;~aLion of the tumor cells (e.g. determination ofthe ~,ession of cell surface molecules). The nucleic acids chosen for transfection can be cl~ i .ed following characterization of the proteins expressed by the tumor cell. ~or instance, ~,res~ion of MHC proteins on the cell surface of the tumor cells and/or expression of the Ii protein in the tumor cell can be assessed. Tumors which express no, or limited amounts of or types of MHC molecules (class I or class II) can be transfected with nucleic acids encoding MHC proteins; tumors which express Ii protein can be transfected with Ii ~nti~en~e sequences. If necessary, following transfection, tumor cells can be screened for introduction of the nucleic acid by using a selectable marker (e.g.
drug resistance) which is introduced into the tumor cells together with the nucleic acid of interest.
Prior to ~tlmini~tration to the subject, the modified tumor cells can be treated to render them incapable of further proliferation in the subject, thereby preventing any possible outgrowth of the modified tumor cells. Possible tre~tment~ include irradiation or mitomycin C treatment, which abrogate the proliferative capacity ofthe tumor cells while m~inl~i,li,lp;
the ability of the tumor cells to trigger antigen-specific and costimulatory signals in T cells and thus to stim~ te an hlllllune response.
The modified tumor cells can be ~lmini~t?red to the subject by injection of the tumor cells into the subject. The route of injection can be, for example, intravenous, intramuscular, intraperitoneal or subcutaneous. ~tlminictration of the modified tumor cells at the site of the original tumor may be beneficial for inducing local T cell-mediated immune responses against the original tumor. Aclmini~tration of the modified tumor cells in a dissemin~tecl manner, e.g. by intravenous injection, may provide systemic anti-tumor immunity and, furthermore, may protect against metastatic spread of tumor cells from the original site. The modified tumor cells can be ~-lmini~tered to a subject prior to or in conjunction with other forms of therapy or can be ~mini~tered after other treatments such as chemotherapy or surgical intervention.
Additionally, more than one type of modified tumor cell can be ~tlministered to a subject. For example, an effective T cell response may require exposure of the T cell to more -W095/03~8 - PCT~S94/084~
2~70~ ~

than one type of costim~ t--ry molecule. Furthermore, the temporal sequence of exposure of the T cell to different costimlll~tory mocules may be important for generating an effective response. For example, it is known that upon activation, a B cell expresses B7-2 early in its response (about 24 hours after stimulation). Subsequently, B7-1 and B7-3 are expressed by the B cell (about 48-72 hours after stim~ tion). Thus, a T cell may require exposure to B7-2 early in the induction of an immlme response by exposure to B7-1 and/or B7-3 in the immune response. Accordingly, different types of modified tumor cells can be ~tlminictered at dirrele~ll times to a subject to generate an effective immune response against the tumor ce~lls.
For example, tumor cells modified to express B7-2 can be ~lmini~tered to a subject.
Following this ~-lmini~tration, a tumor cell from the same tumor but modified to express B7-3 (alone or in conjunction with B7-1) can be ~-lmini~t~red to the subject.
Another method for treating a subject with a tumor is to modify tumor cells in vivo to express B7-2 and/or B7-3, alone or in conjunction with B7-1, MHC molecules and/or an inhibitor of Ii expression. This method can involve modifying tumor cells in vivo by providing nucleic acid encoding the protein(s) to be expressed using vectors and delivery methods effective for in vivo gene therapy as described in a previous section herein.
ltPrn~tively, one or more agents which induce or increase t;x~ies~ion of B7-2 and/or B7-3, and possibly B7-1 or MHC molecules, can be ~rlmini.~t~ted to a subject with a tumor.
The modified tumor cells of the current invention may also be used in a method for preventing or treating metastatic spread of a tumor or ~l~vt;llLing or treating recurrence of a tumor. As demonstrated in detail in one of the following examples, anti-tumor immlmity inclll~e~l by B7-1-~;x~les~ g turnor cells is effective against subsequent challenge by tumor cells, regardless of whether the tumor cells of the re-exposure express B7-1 or not. Thus, ?rltnini~tr~tion of modified tumor cells or modification of tumor cells in vivo as described herein can provide tumor immllnity against cells of the original, unmodified tumor as well as mPt~t~es of the original tumor or possible l~,lOw~l of the original tumor.
The current invention also provides a composition and a method for specifically inducing an anti-tumor response in CD4+ T cells. CD4+ T cells are activated by antigen in conjunction with MHC class II molecules. Association of peptidic fragments of TAAs with MHC class II molecules results in recognition of these antigenic peptides by CD4+ T cells.
Providing a subject with tumor cells which have been modified to express MHC class II
molecules along with B7-2 and/or B7-3, or modified in vivo to express MHC class II
molecules along with B7-2 and/or B7-3, can be useful for directing tumor antigenpresentation to the MHC class II pathway and thereby result in antigen recognition by and activation of CD4+ T cells specific for the tumor cells. Depletion of either CD4+ or CD8+ T
cells in vivo, by ?~mini~tration of anti-CD4 or anti-CD8 antibodies, can be used to Wo 95/03408 ~16 71) ~ 1 PCT/US94/08423 demonstrate that specific anti-tumor immllnity is mediated by a particular (e.g. CD4+) T cell subpopulation.
Subjects initially exposed to modified tumor cells develop an anti-tumor specific T
cell response which is effective against subsequent exposure to unmodified tumor cells. Thus r 5 the subject develops anti-tumor specific imm~lnity. The generalized use of modified tumor cells of the invention from one human subject as an immunogen to induce anti-tumor immlmity in another human subject is prohibited by histocompatibility dirr~ lel1ces between unrelated hnm~n~ However, use of modified tumor cells from one individual to induce anti-turnor immunity in another individual to protect against possible future occurrence of a tumor may be useful in cases of f~mili~l m~ n~ncies. In this situation, the tumor-bearing donor of tumor cells to be modified is closely related to the (non-tumor bearing) recipient of the modified tumor cells and therefore the donor and recipient share MHC antigens. A strong hereditary component has been identified for certain types of m~lign~ncies, for example certain breast and colon cancers. In families with a known susceptibility to a particular m~ n~ncy and in which one individual presently has a tumor, tumor cells from that individual could be modified to express B7-2 and/or B7-3, alone or in combination with B7-1 and ~lmini~tered to susceptible, histocompatible family members to induce an anti-tumor response in the recipient against the type of turnor to which the family is susceptible. This anti-tumor response could provide protective immllnity to subsequent development of a tumor in the immunized recipient.

(15). Tl]mor-Specific T Cell Toler~nce In the case of an experimentally in~ cecl tumor, a subject (e.g. a mouse) can beexposed to the modified tumor cells of the invention before being challenged with unrnodified tumor cells. Thus, the subject is initially exposed to TAA peptides on tumor cells together with B7-2 and/or B7-3, and B7-1 which activates TAA-specific T cells. The activated T cells are then effective against subsequent challenge with unmodified tumor cells.
In the case of a spontaneously arising tumor, as is the case with human subjects, the subject's immune system will be exposed to unmodified tumor cells before exposure to the modified turnor cells of the invention. Thus the subject is initially exposed to TAA peptides on tumor cells in the absence of a costimlll~tory signal. This situation is likely to induce TAA-specific T cell tolerance in those T cells which are exposed to and are in contact with the unmodified tumor cells. Secondary exposure of the subject to modified tumor cells which can trigger a costimulatory signal may not be sufficient to overcome tolerance in TAA-specific T cells which were anergized by primary exposure to the tumor. Use of modified tumor cells to induce anti-tumor immllnity in a subject already exposed to unmodified tumor cells may therefore be most effective in early diagnosed patients with small tumor burdens~ for instance 2~67~

a small localized tumor which has not met~t~i7P~l In this situation, the tumor cells are confined to a limited area of the body and thus only a portion of the T cell repertoire may be exposed to tumor antigens and become anergized. A(lmini~tration of modified tumor cells in a systemic manner, for instance after surgical removal of the localized tumor and 5 modification of isolated tumor cells, may expose non-anergized T cells to tumor antigens together with B7-2 and/or B7-3 alone, or in combination with B7-1 thereby inducing an anti-tumor response in the non-anergized T cells. The anti-tumor response may be effective against possible regrowth of the tumor or against micrometastases of the original tumor which may not have been detected. To overcome widespread peripheral T cell tolerance to 10 tumor cells in a subject, additional signals, such as a cytokine, may need to be provided to the subject together with the modified tumor cells. A cytokine which functions as a T cell growth factor, such as IL-2, could be provided to the subject together with the modified tumor cells. IL-2 has been shown to be capable of restoring the alloantigen-specific responses of previously anergized T cells in an in vitro system when exogenous IL-2 is added atthetimeofsecondaryalloantigenicstim~ tion. Tan,P.,etal.J. Exp. Med. 177,165-173 (1993).
Another approach to ge~G~ g an anti-tumor T cell response in a subject despite tolerance of the subject's T cells to the tumor is to stim~ te an anti-tumor response in T cells from another subject who has not been exposed to the tumor (referred to as a naive donor) 20 and transfer the stimlll~te~l T cells from the naive donor back into the tumor-bearing subject so that the transferred T cells can mount an immune response against the tumor cells. An anti-tumor response is in~ çecl in the T cells from the naive donor by stimulating the T cells in vitro with the modified tumor cells of the invention. Such an adoptive transfer approach is generally prohibited in outbred populations because of histocolllp~libity differences between 25 the transferred T cells and the tumor-bearing recipient. However, advances in allogeneic bone marrow transplantation can be applied to this situation to allow for acceptance by the recipient of the adoptively transferred cells and prevention of graft versus host disease. First, a tumor-bearing subject (referred to as the host) is prepared for and receives an allogeneic bone marrow transplant from a naive donor by a known procedure. Preparation of the host 30 involves whole body irradiation, which destroys the host's immune system, including T cells tolerized to the tumor, as well as the tumor cells themselves. Bone marrow transplantation is accompanied by treatment(s) to prevent graft versus host disease such as depletion of mature T cells from the bone marrow graft, treatment of the host with immlln-~suppressive drugs or treatment of the host with an agent, such as CTLA4Ig, to induce donor T cell tolerance to 35 host tissues. Next, to provide anti-tumor specific T cells to the host which can respond against residual tumor cells in the host or regrowth or met~t~ces of the original tumor in the host, T cells from the naive donor are ~timlll~te~l in vitro with tumor cells from the host WO 95/03408 21 ~ 7 0 91 PCT/US94/08423 which have been modified, as described herein, to express B7-2 and/or B7-3. Thus, the donor T cells are initially exposed to tumor cells together with a costim~ tory signal and therefore are activated to respond to the tumor cells. These activated anti-tumor specific T
cells are then transferred to the host where they are reactive against unmodified tumor cells.
5 Since the host has been reconstituted with the donor's immllne system, the host will not reject the transferred T cells and, additionally, the tre~tm~nt of the host to prevent graft versus host disease will prevent reactivity of the transferred T cells with normal host tissues.

H. Admini~.tration of Thera~eutic Forrn~ of R ~ ~ymphocyte Anti~en~
The peptides ofthe invention are allmini~ered to subjects in a biologically compatible form suitable for ph~rm~reutical atlministration in vivo to either enhance or suppress T cell mediated immlme response. By "biologically compatible form suitable for a-lministration in vivo" is meant a form of the protein to be ~lministered in which any toxic effects are outweighed by the therapeutic effects of the protein. The term subject is intended to include 15 living org~ni~ms in which an imml~ne response can be elicited, e.g., m~mm~l~. Examples of subjects include hnm~n~, dogs, cats, mice, rats, and transgenic species thereof.A-lministration of a peptide having the activity of a novel B lymphocyte antigen as described herein can be in any ph~rm~rological form including a thc.d~ulically active amount of peptide alone or in combination with a peptide having the activity of another B lymphocyte 20 antigen and a ph~rm~c~eutically acceptable carrier. A-lministration of a therapeutically active amount of the therapeutic compositions of the present invention is defined as an amount effective, at dosages and for periods of time n~ces~ay to achieve the desired result. For example, a therapeutically active arnount of a peptide having B7-2 activity may vary according to factors such as the disease state, age, sex, and weight of the individual, and the 25 ability of peptide to elicit a desired response in the individual. Dosage regima may be adjusted to provide the optimum th~l~eulic response. For example, several divided doses may be atlminiet~?red daily or the dose may be ~.ropolLionally reduced as indicated by the exigencies of the therapeutic situation.
The active compound (e.g., peptide) may be ~tlministered in a convenient manner 30 such as by imjection (subcutaneous, intravenous, etc.), oral ~rlministration, inhalation, transdermal application, or rectal a~lministration. Depending on the route of atlministration, the active compound may be coated in a material to protect the compound from the action of enzymes, acids and other natural conditions which may inactivate the compound.
To admini~ter a peptide having B7-2 activity by other than parenteral a~lministration, 35 it may be necessary to coat the peptide with, or co-a~lminister the peptide with, a material to prevent its inactivation. For example, a peptide hving B7-2 activity may be a-lmini~tered to an individual in an ~ ;ate carrier, diluent or adjuvant, co-a~lministered with enzyme wo 95,03408 ~ ~ 6 '~ O ~ ~ PCT/US94/08423 inhibitors or in an ~propl;ate carrier such as liposomes. Ph~rm~ceutically acceptable diluents include saline and aqueous buffer solutions. Adjuvant is used in its broadest sense and includes any immllne stimulating compound such as interferon. Adjuvants contemplated herein include resorcinols, non-ionic surfactants such as polyoxyethylene oleyl ether and n-S he~ cyl polyethylene ether. Enzyme inhibitors include pancreatic trypsin inhibitor,diisopropylfluorophosphate (DEP) and trasylol. Liposomes include water-in-oil-in-water emulsions as well as conventional liposomes (Strejan et ~L, (1984) J. Neuroimmunol 1:27).
The active compound may also be ~imini~tered parenterally or intraperitoneally.
Dispersions can also be prepared in glycerol, liquid polyethylene glycols, and mixtures l 0 thereof and in oils. Under ordinary conditions of storage and use, these preparations may contain a preservative to prevent the growth of microorg~ni~m~
Pharmaceutical compositions suitable for injectable use include sterile aqueous solutions (where water soluble) or dispersions and sterile powders for the extemporaneous aldlion of sterile injectable solutions or dispersion. In all cases, the composition must be l 5 sterile and must be fluid to the extent that easy syringability exists. It must be stable under the conditions of manufacture and storage and must be preserved against the cont~min~ting action of microor~ni~m~ such as bacteria and fungi. The carrier can be a solvent or dispersion medium cont~inin~, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyetheylene glycol, and the like), and suitable mixtures 20 thereof. The proper fluidity can be ~ ed, for example, by the use of a coating such as lecithin, by the m~inlen~nce of the required particle size in the case of dispersion and by the use of sllrf~ct~nt~ Prevention of the action of microorg~ni~m~ can be achieved by various antibacterial and antifungal agents, for example, parabens, chlorobutanol, phenol, asorbic acid, thimerosal, and the like. In many cases, it will be preferable to include isotonic agents, 25 for example, sugars, polyalcohols such as manitol, sorbitol, sodium chloride in the composition. Prolonged absorption of the injectable compositions can be brought about by including in the composition an agent which delays absorption, for example, alnminllm monostearate and gelatin.
Sterile injectable solutions can be prepared by incorporating active compound (e.g., 30 peptide having B7-2 activity) in the required amount in an appr~,pliate solvent with one or a combination of ingredients enumerated above, as required, followed by filtered sterilization.
Generally, dispersions are prepared by incorporating the active compound into a sterile vehicle which contains a basic dispersion medium and the required other ingredients from those enumerated above. In the case of sterile powders for the pl~d~ion of sterile 35 injectable solutions, the preferred methods of ~l~paldLion are vacuum drying and freeze-drying which yields a powder of the active ingredient (e.g., peptide) plus any additional desired ingredient from a previously sterile-filtered solution thereof.

9~/0340~ ~16 ~ O 91 PCTIUS94l08423 When the active compound is suitably protected? as described above, the protein may be orally ~tlmini~tered, for example, with an inert diluent or an ~imil~ble edible carrier. As used herein "ph~rrn~eutically acceptable carrier" includes any and all solvents? dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents?
and the like. The use of such media and agents for ph~rm~ceutically active substances is well known in the art. Except insofar as any conventional media or agent is incompatible with the active compound, use thereof in the therapeutic compositions is contemplated.
Supplement~ry active compounds can also be incorporated into the compositions.
It is especially advantageous to formulate palel1teldl compositions in dosage unit form for ease of ~-lmini~tration and uniformity of dosage. Dosage unit form as used herein refers to physically discrete units suited as unitary dosages for the m~mm~ n subjects to be treated; each unit cont~inin~ a pre.letermined quantity of active compound calculated to produce the desired therapeutic effect in association with the required ph~rm~reutical carrier.
The specification for the dosage unit forms of the invention are dictated by and directly dependent on (a) the unique characteristics of the active compound and the particular therapeutic effect to be achieved, and (b) the limitations inherent in the art of compounding such an active compound for the tre~tment of sensitivity in individuals.

I. Identification of Cytokines ~nduced by Costimulation The nucleic acid sequences encoding peptides having the activity of novel B
Iymphocyte antigens as described herein can be used to identify cytokines which are produced by T cells in response to stimulation by a form of B lymphocyte antigen, e.g., B7-2.
T cells can be suboptimally stim~ te~i in vitro with a primary activation signal, such as phorbol ester, anti-CD3 antibody or preferably antigen in association with an MHC class II
molecule, and given a costimulatory signal by a stim~ tQry form of B7-2 antigen, for in.~t~n~e by a cell transfected with nucleic acid encoding a peptide having B7-2 activity and expressing the peptide on its surface or by a soluble, stimulatory form of the peptide. Known cytokines released into the media can be identified by ELISA or by the ability of an antibody which blocks the cytokine to inhibit T cell proliferation or proliferation of other cell types that is in~llced by the cytokine. An IL-4 ELISA kit is available from Genzyme (Cambridge MA~, as is an IL-7 blocking antibody. Blocking antibodies against IL-9 and IL-12 are available from Genetics Institute (Cambridge, MA).
An in vitro T cell costimulation assay as described above can also be used in a method for identifying novel cytokines which may be in~ ced by costim~ tion. If a particular activity in~ ce~l upon costimlll~tion, e.g., T cell proliferation, cannot be inhibited by addition of blocking antibodies to known cytokines, the activity may result from the action of an wo 95,03408 2~6 ~ PCT/US94/08423 unkown cytokine. Following costimulation, this cytokine could be purified from the media by conventional methods and its activity measured by its ability to induce T cell proliferation.
To identify cytokines which prevent the induction of tolerance, an in vitro T cell costimulation assay as described above can be used. In this case, T cells would be given the primary activation signal and contacted with a selected cytokine, but would not be given the cosfim~ tory signal. After washing and resting the T cells, the cells would be rechallenged with both a primary activation signal and a costim~ tory signal. If the T cells do not respond (e.g., proliferate or produce IL-2) they have become tolerized and the cytokine has not prevented the induction of tolerance. However, if the T cells respond, induction of tolerance has been prevented by the cytokine. Those cytokines which are capable of preventing the induction of tolerance can be targeted for blockage in vivo in conjunction with reagents which block B lymphocyte antigens as a more efficient means to induce tolerance in transplant recipients or subjects with autoimml-ne diseases. For example, one could ~mini~ter a B7-2 blocking reagent together with a cytokine blocking antibody to a subject.
J. Identification of Molecules which Inhihit Costimulation Another application of the peptide having the activity of a novel B lymphocyte antigen of the invention (e.g., B7-2 and B7-3) is the use of one or more of these peptides in screening assays to discover as yet undefined molecules which are inhibitors of costimulatory ligand binding and/or of intracellular ~i~n~lin~ through T cells following costiml-l~tion. For example, a solid-phase binding assay using a peptide having the activity of a B lymphocyte antigen, such as B7-2, could be used to identify molecules which inhibit binding of the antigen with the a~plopliate T cell ligand (e.g., CTLA4, CD28). In addition, an in vitro T
cell costim~ tion assay as described above could be used to identify molecules which interfere with intracellular ~ipn~ling through the T cells following costimlll~tion as cleterminPcl by the ability of these molecules to inhibit T cell proliferation and/or cytokine production (yet which do not prevent binding of B lymphocyte antigens to their receptors).
For example, the compound cyclosporine A inhibits T cell activation through stimlll~tion via the T cell receptor pathway but not via the CD28/CTLA4 pathway. Therefore, a different intracellular si~n~ling pathway is involved in costimulation. Molecules which interfere with intracellular sign~lin~ via the CD28/CTLA4 pathway may be effective as immunosuppressive agents in vivo (similar to the effects of cyclosporine A).

K. Identification of Molecules which Modulate B Lymphocyte Anti~en l~xpression The monoclonal antibodies produced using the proteins and peptides of the current invention can be used in a screening assay for molecules which modulate the expression of B
lymphocyte antigens on cells. For example, molecules which effect intracellular sign~ling WO 95/03408 ~ ~. 6 ~ ~ ~1 PCT/US94/08423 .

which leads to induction of B Iymphocyte antigens, e.g. B7-2 or B7-3, can be identified by assaying e~lession of one or more B lymphocyte antigens on the cell surface. ~çcl~lce~l - imml]n~fluorescent staining by an anti-B7-2 antibody in the presence of the molecule would indicate that the molecule inhibits intracellular signals. Molecules which upregulate B
lymphocyte antigen expression result in an increased immunofluorescent st~ining.Alternatively, the effect of a molecule on expression of a B lymphocyte antigen, such as B7-2, can be determined by detecting cellular B7-2 mRNA levels using a B7-2 cDNA as a probe For example, a cell which expresses a peptide having B7-2 activity can be contacted with a molecule to be tested, and an increase or decrease in B7-2 mRNA levels in the cell detected by standard technique, such as Northern hybridization analysis or conventional dot blot of mRNA or total poly(A+)RNAs using a B7-2 cDNA probe labeled with a detectable marker.
Molecules which modulate B lymphocyte antigen expression may be useful therapeutically for either upregulating or downregulating immune responses alone or in conjunction with soluble blocking or stimulating reagents. F~r instance, a molecule which inhibits expression of B7-2 could be ~lmini~tered together with a B7-2 blocking reagent for immunosuppressive purposes. Molecules which can be tested in the above-described assays include cytokines such as IL-4, yINF, IL-10, IL-12, GM-CSF and prost~gl~lin~
This invention is further illustrated by the following examples which should not be construed as limiting. The contents of all references and published patent applications cited throughout this application are hereby incorporated by reference.

The following methodology was used in Examples 1, 2 and 3.

METHODS AND MATERIALS
A Cells Mononuclear cells were isolated by Ficoll-Hypaque density gradient centrifugation from single cell suspensions of normal human spleens and were separated into E- and E+
fractions by rosetting with sheep red blood cells (Boyd, A.W., et al. (1985) J. Immunol. 134, 1516). B cells were purified from the E- fraction by adherence of monocytes on plastic and depletion of residual T, natural killer cells (NK) and residual monocytes by two treatments with anti-MsIgG and anti-MsIgM coated magnetic beads (Advanced Magnetics, Cambridge, MA), using monoclonal antibodies: anti-CD4, -CD8, -CDl lb, -CD14 and -CD16. CD4+ T
cells were isolated from the E+ fraction of the same spleens after adherence on plastic and depletion of NK, B cells and residual monocytes with magnetic beads and monoclonal antibodies: anti-CD20, -CDl lb, -CD8 and -CD16. CD28+ T cells were identically isolated from the E+ fraction using anti-CD20, -CD 1 l b, -CD 14 and -CD 16 monoclonal antibodies.
The efficiency of the purification was analyzed by indirect immunofluorescence and flow ~7~

cytometry using an EPICS flow cytometer (Coulter). B cell ~ ualdliOnS were >95% CD20+, <2% CD3+, <1% CD14+. CD4+ T cell preparations were >98% CD3+, >98% CD4+.<1%
CD8+, <1% CD20+, <1% CD14+. CD28+ T cell preparations were >98% CD3+, >98%
CD28+, <1% CD20+, <1% CD14+.
R. Monoclonal Antibodies and Fusion Proteins Monoclonal antibodies were used as purified Ig unless in~lir~ted otherwise: anti-B7:133, IgM is a blocking antibody and has been previously described (FreeAm~n, A.S. et al.
(1987) Immunol. 137, 3260-3267); anti-B7:Bl.l, IgGl (RepliGen Corp., Cambridge, MA) (Nickoloff, B., et al (1993) Am. J. Pathol. 142, 1029-1040) is a non-blocking monoclonal antibody; BB-1: IgM is a blocking antibody (Dr. E. Clark, University of Washington, Seattle, WA) (Yokochi, T., et al. (1982) J. Immunol. 128, 823-827); anti-CD20: B1, IgG2a (St~henk~, P., et al.(1980) J. Immunol. L~, 1678-1685); anti-BS: IgM (Freerln~n, A., et al.
(1985) J. Immunol. 134, 2228-2235); anti-CD8: 7PT 3F9, IgG2a; anti-CD4: l9ThySD7, IgG2a; anti-CDl lb: Mol, IgM and anti-CD14: Mo2, IgM (Todd, R, et al. (1981) J. Immunol.
126, 1435-1442); anti-MHC class II: 9-49, IgG2a (Dr R. Todd, University of Michigan, Ann Arbor) (Todd, R.I., et al. (1984) Hum Immunol. 10, 23-40; anti-CD28: 9.3, IgG2a (Dr. C.
June, Naval Research Institute, Bethesda) (Hansen, J.A., et al. (1980) Immunogenetics. 10, 247-260); anti-CD16: 3G8, IgGl (used as ascites) (Dr. J. Ritz, Dana-Farber Cancer Tn~titllte, Boston); anti-CD3: OKT3, IgG2a hybridoma was obtained from the American Type Culture Collection and the purified monoclonal antibody was adhered on plastic plates at a concentration of 1,ug/ml; anti-CD28 Fab fr~gment~ were generated from the 9.3 monoclonal antibody, by papain digestion and purification on a protein A column, according to the manufacturer's instructions (Pierce, Rockford, IL). Human CTLA4 fusion protein (CTLA4Ig) and control fusion protein (control-Ig) were prepared as previously described (Gimmi, C.D., et al. (1993) Proc. Natl. Acad. Sci USA ~1:6586-6590); Boussiotis, V., et al J.
Exp. Med. (accepted for publication)).

C. CHO Cell Tr~n~fection B7-1 transfectants (CHO-B7) were prepared from the B7-1 negative chinese hamsterovary (CHO) cell line, fixed with paraformaldehyde and used as previously described (Gimmi, C.D., et al. Proc. Natl. Acad. Sci USA ~, 6575-6579).

D. In Vitro B Cell Activation and Selection of B7+ and B7- Cells Splenic B cells were cultured at 2X106 cells/ml in complete culture media, {RPMI1640 with 10% heat inactivated fetal calf serum (FCS), 2mM gh1t~nnin~, 1 mM sodium pyruvate, penicillin (100 units/ml), streptomycin sulfate (100~1g/ml) and gentamycin sulfate -~ o 95,03408 2 l 6 ~ O 9 1 PCT/US94/08423 (5~1g/ml)}, in tissue culture flasks and were activated by cro.~.~linking of sIg with affinity purif1ed rabbit anti-human IgM coupled to Affi-Gel 702 beads (Bio-Rad), Richmond, CA) (Boyd, A.W., et al., (1985) J: Immunol. 134,1516) or by cro.s.~linking of MHC class II with 9-~9 antibody coupled to Affi-Gel 702 beads. B cells activated for 72 hours, were used as total - 5 activated B cell populations or were indirectly stained with anti-B7 (B 1.1) monoclonal antibody and fluorscein isothiocyanate (FITC) labeled goat anti-mouse immunoglobulin (Fisher, Pittsburgh, PA), and fractionated into B7-1+ and B7-1- populations by flow cytometric cell sorting (EPICS Elite flow cytometer, Coulter).

F. Tmmlmoflouoresce~ce ~ntl Flow Cytornetrv For surface phenotype analysis populations of B cells activated by either slg or MHC
class II cro.~slinking for 6, 12, 24, 48, 72 and 96 hours were stained with either anti-B7 (133), BB-1 monoclonal antibodies, control IgM antibody, CTLA4Ig or control-Ig. Cell suspensions were stained by two step indirect membrane st~inin~ with l O~lg/ml of primary monoclonal antibody followed by the a~l)ropliate secondary reagents. Specifically, immunoreactivity with anti-B7 (133) and BB-1 monoclonal antibodies was studied by indirect staining using goat anti-mouse Ig or immlmc)globulin FITC (Fisher) as secondary reagent and immlm~reactivity with fusion proteins was studied using biotinylated CTLA4Ig or biotinylated control-Ig and streptavidin-phycoerythrin as secondary reagent. PBS
cont~ining 10% AB serum was used as diluent and wash media. Cells were fixed with 0.1 %
p~dru,..laldehyde and analyzed on a flow cytometer (EPICS Elite Coulter).

F. Prolifer~tion A~
T cells were cultured at a concentration of lxl O5 cells per well in 96-well flat bottom microtiter plate at 37C for 3 days in 5% CO2. Syngeneic activated B cells (total B cell population or B7+ and B7- fractions) were irr~ te-l (2500 rad) and added into the cultures at a concentration of 1 x 105 cells per well. Factors under study were added to the required concentration for a total final volume of 200 Ill per well. When indicated, T cells were incubated with anti-CD28 Fab (final concentration of lO,ug/ml), for 30 minutes at 4C, prior to addition in c;~ nental plates. Similarly, CHO-B7 or B cells were incubated with CTL~4Ig or control-Ig (lO~lg/ml) for 30 minllt~s at 4C. Thymidine incorporation as an index of mitogenic activity, was assessed after incubation with 1 ~lCi (37kBq) of {methyl-3H}
thymidine (Du Pont, Boston, MA) for the last 15 hours of the culture. The cells were harvested onto filters and the radioactivity on the dried filters was measured in a Pharrnacia beta plate liquid scintilation counter.

~16~

G. JT -2 ~n(l IL-4 A~ay IL-2 and IL-4 concenkations were assayed by ELISA (R&D Systems, Minneapolis, MN and BioSource, Camarillo, CA) in culture supern~t~nt~ collected at 24 hours after initiation of the culture.

li ~pre~sion of a Novel CTT ,~4 ~.~and on Activated B Cells Whi-`h Induces T Cell Proliferation Since crosslinking surface Ig in~l~ce~ human resting B cells to express B7-l maximally (50-80%) at 72 hours, the ability of activated human B lymphocytes to induce submitogenically activated T cells to proliferate and secrete IL-2 was determined. Figure 1 depicts the costimulatory response of human splenic CD28+ T cells, submitogenically activated with anti-CD3 monoclonal antibody, to either B7 (B7-1) transfected CHO cells (CHO-B7) or syngeneic splenic B cells activated with anti-Ig for 72 hours. 3H-Thymidine incorporation was ~sessed for the last 15 hours of a 72 hours culture. IL-2 was assessed by ELISA in supern~t~nt~ after 24 hours of culture (Detection limits of the assay: 31-2000 pg/ml). Figure 1 is .~;~.ese~ e of seventeen experiments.
Submitogenically activated CD28+ T cells proliferated and secreted high levels of IL-2 in response to B7-1 costim~ ion provided by CHO-B7 (Figure 1, panel a). Both proliferation and IL-2 secretion were totally inhibited by blocking the B7-1 molecule on CHO cells with either anti-B7- 1 monoclonal antibody or by a fusion protein for its high affinity receptor, CTLA4. Similarly, proliferation and IL-2 secretion were abrogated by blocking B7-1 ~ign~lling via CD28 with Fab anti-CD28 monoclonal antibody. Control monoclonal antibody or control fusion protein had no effect. Nearly identical costimlll~tion of proliferation and IL-2 secretion was provided by splenic B cells activated with anti-Ig for 72 hours (panel b). Though anti-B7-1 monoclonal antibody could completely abrogate both proliferation and IL-2 secretion delivered by CHO-B7, anti-B7-1 monoclonal antibody con~i~tently inhibited proliferation in~ ce~l by activated B cells by only 50% whereas IL-2 secretion was totally inhibited. In contrast to the partial blockage of proliferation in~l~lcecl by anti-B7-1 monoclonal antibody, both CTLA4Ig and Fab anti-CD28 monoclonal antibody completely blocked proliferation and IL-2 secretion. These results are consistent with the hypothesis that activated human B cells express one or more additional CTLA4/CD28 ligands which can induce T cell proliferation and IL-2 secretion.

WO g~/~3408 21~ 7 ~ 91 PCTIUS94/08423 .

aled ~um~n Splenic B Celle F.~press CT~ ~4 T ~nd(s) I)ietinct from n7-1 In light of the above observations, whether other CTLA4 binding cour;ter-receptors - 5 were ex~ressed on activated B cells was determined. To this end, human splenic B cells were activated for 72 hours with anti-Ig and then stained with an anti-B7-1 monoclonal antibody (Bl.1) which does not inhibit B7-1 mediated costim~ tion. Fluoroscein isothiocyanate (FITC) and mAb B 1.1 were used with flow cytometric cell sorting to isolate B7- 1 + and B7- 1 ~
fractions. The resulting post-sort positive population was 99% B7-1+ and the post-sort negative population was 98% B7- 1 ~ (Figure 2).
To ~mine the costimulatory potential of each population, human splenic CD28+ T
cells were submitogenically stimulated with anti-CD3 monoclonal antibody in the presence of irradiated B7-1+ or B7-1- anti-Ig activated (72 hours) splenic B cells. 3H-Thymidine incorporation was assessed for the last 15 hours of a 72 hours culture. IL-2 was assessed by ELISA in supernatants after 24 hours of culture (Detection limits of the assay: 31 -2000 pg/ml). The results of Figure 3 are representative of ten experiments. B7-1+ B cells in~ recl anti-CD3 activated T cells to proliferate and secrete IL-2 (Figure 3a) but not IL-4. As was observed with the unfractionated activated B cell population, anti-B7-1 monoclonal antibody (133) inhibited proliferation only 50% but con.ei.ett?ntly abrogated IL-2 secretion. As above, CTLA4Ig binding or blockade of CD28 with Fab anti-CD28 monoclonal antibody completely inhibited both proliferation and IL-2 secretion. Control monoclonal antibody and control-Ig were not inhibitory. In an attempt to identify other potential CTLA4/CD28 binding costimulatory ligand(s) which might account for the residual, non-B7 mediated proliferation delivered by B7+ B cells, the effect of BB-1 monoclonal antibody on proliferation and IL-2 secretion was exarnined. As seen, BB-l monoclonal antibody completely inhibited both proliferation and IL-2 secretion (Figure 3a). Figure 3b displays the costimulatory potential of B7-1- activated human splenic B cells. Irradiated B7-1- activated (72 hr) B cells could also deliver a significant costimulatory signal to submitogenically activated CD4+ lymphocytes.
This costim~ tion was not accompanied by detectable IL-2 (Figure 3b) or IL-4 accumulation and anti-B7-1 monoclonal antibody did not inhibit proliferation. However, CTLA4Ig, Fab anti-CD28 monoclonal antibody, and BB-l monoclonal antibody all completely inhibited proliferation.
Phenotypic analysis ofthe B7-1+ and B7-1- activated splenic B cells confirmed the above functional results. Figure 4 shows the cell surface expression of B7-1, B7-2 and B7-3 on fractionated B7-1+ and B7-1- activated B cell. As seen in Figure 4, B7-1+ activated splenic B cells stained with anti-B7-1 (133) monoclonal antibody, BB-1 monoclonal antibody, and bound CTLA4-Ig. In contrast, B7- activated splenic B cells did not stain with æ~6~9~ -- anti-B7-1 (133) monoclonal antibody but did stain with BB-l monoclonal antibody and CTLA4Ig. These phenotypic and functional results demonstrate that both B7-1+ and B7-1-activated (72 hours) human B Iymphocytes express CTLA4 binding counter-receptor(s) which: 1) can induce submitogenically activated T cells to proliferate without detectable IL-2 secretion; and 2) are identified by the BB-l monoclonal antibody but not anti-B7-1 monoclonal antibody. Thus, these CTLA4/CD28 ligands can be distinguished on the basis of their temporal expression after B cell activation and their reactivity with CTLA41g and anti-B7 monoclonal antibodies. The results of Figure 4 are representative of five experiments.

Three I)istinct CT~,~4/CD28 1 i~nds Are Fx~ressed Followin~ Human B CellActivation To fletçrmine the sequential ~x~les~ion of CTLA4 binding counter-receptors following activation, human splenic B cells were activated by crosslinking of either surface Ig or MHC class II and the expression of B7-1, B7-3 and B7-2 binding proteins were examined by flow cytometric analysis. Ig or MHC class II cro~linkinp in~ ce~l a similar pattern of CTLA4Ig binding (Figures 5 and 6). Figure S is representative of the results of 25 experiments for anti-B7-1 and BB-l binding and 5 experiments for CTLA4Ig binding.
Figure 6 is ~ ,resellL~Li~e of 25 experiments for anti-B7-1 binding and 5 experiments for CTLA4Ig binding. The results of these experiments indictes that prior to 24 hours, none of these molecules are expressed. At 24 hours post-activation, the majority of cells express a protein that binds CTLA4Ig (B7-2), however, fewer than 20% express either B7-1 or B7-3.
Crosslinkin~ of MHC class II induces m~im~l ~x~l~s~.ion and intensity of B7-1 and B7-3 at 48 hours whereas cros~linkin~ of Ig induces maximal ~ ression at 72 hours and ex~les~ion declines thereafter. These results suggest that an additional CTLA4 binding counter-receptor is expressed by 24 hours and that the temporal expression of the distinct B7- 1 and B7-3 proteins appears to coincide.
A series of experiments was conducted to determine whether the temporal expression of CTLA4 binding counter-receptors differentially correlated with their ability to costimnl~te T cell proliferation and/or IL-2 secretion. Human splenic CD28+ T cells submitogenically ~tim~ te~l with anti-CD3 were cultured for 72 hours in the presence of irradiated hurnan splenic B cells that had been previously activated in vitro by sIg crosslinkin~ for 24, 48, or 72 hours. IL-2 secretion was ~es~ed by ELISA in supernatants after 24 hours and T cell proliferation as ~sesse~l by 3H-thymidine incorporation for the last 15 hours of a 72 hour culture. The results of Figure 7 are representative of 5 experiments. As seen in Figure 7a, 24 hour activated B cells provided a costim~ tory signal which was accompanied by modest ~ - -WO 95to3~ 7 Q ~1 PCT/US94/08423 levels o:~ IL-2 production, although the m~itn-le of proliferation was significantly less than observed with 48 and 72 hours activated human B cells (note differences in scale for 3H-Thymidine incorporation). Neither proliferation nor IL-2 accurnulation was inhibited by anti-B7-1 (133) or BB-l. In contrast, with CTLA4Ig and anti-CD28 Fab monoclonal 5 antibody totally abrogated proliferation and IL-2 accumulation. B cells activated for 48 hours, provided costimulation which resulted in nearly maximal proliferation and IL-2 secretion (Figure 7b). Here, anti-B7-1 (133) monoclonal antibody, inhibited proliferation approximately 50% but totally blocked IL-2 accu~nulation. BB-l monoclonal antibody totally inhibited both proliferation and IL-2 secretion. As above, CTLA4Ig and Fab 10 anti-CDZ8 also totally blocked proliferation and IL-2 production. Finally, 72 hour activated B cells intluced T cell response identical to that in(ll1ce~1 by 48 hour activated B cells. Similar results are observed if the submitogenic signal is delivered by phorbol myristic acid (PMA) and if the human splenic B cells are activated by MHC class II rather than Ig cro.~linking.
These results indicate that there are three CTLA4 binding molecules that are temporarily 15 ~ ssed on activated B cells and each can induce submitogenically stimulated T cells to proliferate. Two of these molecules, the early CTLA4 binding counter-receptor (B7-2) and B7-1 (133) induce IB-2 production whereas B7-3 inf31l~çs proliferation without detectable IL-2 production.
Previous studies provided conflicting evidence whether the anti-B7 monoclonal 20 antibody,l33 and monoclonal antibody BB-l identified the same molecule (Free~lm~n, A.S.
et al. (1987) Immunol. ~, 3260-3267; Yokochi, T., et al. (1982) J: Immunol. 128, 823-827;
Freeman, G.J., et al. (1989) ~ Immunol. 1~, 2714-2722.). Although both monoclonal antibodies identified molecules expressed 48 hours following human B-cell activation, several reports suggested that B7 (B7-1) and the molecule identified by monoclonal antibody 25 BB-1 were distinct since they were differentially expressed on cell lines and B cell neoplasms (Free~im~n, A.S. et al. (1987) Immunol. 137, 3260-3267; Yokochi, T., et al. (1982) J.
Immunol. 128, 823-827; Freeman, G.J., et al. (1989) J. Immunol. 143, 2714-2722; Clark, E
and Yokochi, T. (1984) Leulcocyte Typing, Ist International References Workshop. 339-346;
Clark, E., et al. (1984) Leukocyte Typing, 1stInternational References Workshop. 740). In 30 addition, immllnf~precipitation and Western Blotting with these IgM monoclonal antibodies suggested that they identified different molecules (Clark, E and Yokochi, T. (1984) Leukocyte Typing, Ist International References Workshop. 339-346; Clark, E., et al. (1984) Leukocyte Typing, I st International References Workshop. 740). The original anti-B7 monoclonal antibody, 133, was generated by immlmi7~tion with anti-immunoglobulin35 activated human B lymphocytes whereas the BB-l monoclonal antibody was generated by imml-ni7~tif~n with a baboon cell line. Thus, the BB-I monoclonal antibody must identify an epitope on human cells that is conserved between baboons and hllm~n~

21~7~ 68-Following the molecular cloning and expression of the human B7 gene (B7-1), B7 transfected COS cells were found to be identically stained with the anti-B7 (133) and BB-1 monoclonal antibodies and that they both ple~ led the identical broad molecular band (44-54kD) strongly suggesting that they identified the same molecule (Freeman, G.J., et al.
(1989) J. Immunol. 143, 2714-2722). This observation was unexpected since the gene encoding the molecule identified by the BB-1 monoclonal antibody had been previously mapped to chromosome 12 (Katz, ~.E., et al. (1985) Eur. J. Immunol. 103-6), whereas the B7 gene was located by two groups on chromosome 3 (Freeman, G.J., et al. (1992) Blood. 79, 489-494; Selvakumar, A., et al. (1992) Immunogenetics 36, 175-181.). Subsequently, additional discrepancies between the phenotypic expression of B7 (B7-1) and the molecule identified by the BB-1 monclonal antibody were noted. BB-l monoclonal antibody stained thymic epithelial cells (Turka, L.A., et al. (1991) J. Immunol. 1~, 1428-36; Munro, J.M., et al. Blood submitted.) and keratinocytes (Nickoloff, B., et al (1993) ~m. J. Pathol. 142, 1029-1040; Augustin, M., et al. (1993) J. Invest. Dermatol. 100, 275-281.) whereas anti-B7 did not. Recently, Nickoloff et al. (1993) ~lm. J. Pathol. ~, 1029-1040, reported discordant expression of the molecule identified by the BB-l monoclonal antibody and B7 on keratinocytes using a BB-1 and anti-B7 (B 1.1 and 133) monoclonal antibodies. Nickoloff et al. also demonstrated that these BB-l positive cells did not express B7 mRNA yet bound CD28 transfected COS cells providing further support for the existence of a distinct protein which binds monoclonal antibody BB-l.
The present finrlin~ confirm that there is an additional CTLA4 counter-receptor identified by the BB-l monoclonal antibody, B7-3, and that this protein appears to be functionally distinct from B7-1 (133). Although the ~ ression of B7-1 and B7-3 following B cell activation appears to be concordant on B7 positive B cells, these studies demonstrate that the B7-3 molecule is also expressed on B7 negative activated B cells. More importantly, the B7-3 molecule appears to be capable of inducing T cell proliferation without detectable IL-2 or IL-4 production. This result is similar to the previous observation that ICAM-I could costim~ t~ T cell proliferation without detectable IL-2 or IL-4 production (Boussiotis, V., et al J. E~p. Med. (accepted for publication)). These data indicate that the BB-l monoclonal antibody recognizes an epitope on the B7-1 protein and that this epitope is also found on a distinct B7-3 protein, which also has costimulatory function. Phenotypic and blocking studies demonstrate that the BB-1 monoclonal antibody could detect one (on B7 negative cells) or both (on B7 positive cells) of these proteins. In conkast, the anti-B7 monoclonal antibodies, 133 and Bl.l detect only the B7-1 protein. Taken together, these results suggest that by 48 hours post B-cell activation by cros~linking of surface immlmt~globulin or MHC
class II, B cells express at least two distinct CTLA4 binding counter-receptors, one identified ~ O 95/03408 21 6 7 ~ 9 ~ PCT/US94/08423 by both anti-B7 and BB-1 monclonal antibodies and the other identified only by BB-l monoclonal antibody.
The B7-2 antigen is not detectable on activated B cells after 12 hours, but by 24 hours it is strongly expressed and functional. This molecule appears to signal via CD28 since - 5 proliferation and IL-2 production are completely blocked by Fab anti-CD28 monoclonal antibody. At 48 hours post activation, IL-2 secretion seems to be accounted for by B7-1 costimulation, since anti-B7 monoclonal antibody completely inhibits IL-2 production.
Previous studies and results presented here demonstrate that B7 (B7-1) is neither expressed (Free~lm~n, A.S. et al. (1987) Immunol. ~1, 3260-3267; Free~lm~n, A.S., et al.
(1991) Cell. Immunol. 137, 429-437) nor capable of costimulating T cell proliferation or IL-2 secretion until 48 hours post B-cell activation. Previous studies have shown that activation of T cells via the TCR in the absence of cos~im~ tion (Gimmi, C.D., et al. (1993) Proc. Natl.
Acad. Sci USA 90:6586-6590; Schwartz, R.H., et al. (1989) Cold Spring Harb. Symp. Quant.
Biol 54, 605-10; Beverly, B., et al. (1992) Int. Immunol. 4. 661-671.) and lack of IL-2 (Boussiotis, V., et al J. ~p. Med. (submitted); Beverly, B., et al. (1992) Int. Immunol. _, 661-671; Wood, M., et al. (1993) J. Exp. Med. ~11, 597-603) results in anergy. If B7-1 were the only costimulatory molecule capable of inducing IL-2 secretion, T cells would be anergized within the first 24 hours following activation since there is no B7-1 present to costimulate IL-2 production. Therefore, the existence of another, early inducible costimulatory molecule, which can costim~ te IL-2 secretion during the first 24 hours would be necessary to induce an effective immune response rather than anergy. The appearance of the early CTLA4 binding counter-receptor, B7-2, between 12 and 24 hours post B cell activation, fulfills this function.
Two observations shed light on the biologic and potential clinical significance of these two additional CTLA4 binding counter-receptors. First, B7 (B7-1) deficient mouse has been developed and its antigen ~lese~ g cells were found to still bind CTLA4Ig (Freeman and Sharpe manuscript in plc;î,aldlion). This mouse is viable and isolated mononuclear cells induce ~etect~hle levels of IL-2 when cultured with T cells in vitro. Therefore, an alternative CD28 co~im~ tory counter-receptor or an alternative IL-2 producing pathway must be functional. Second, thus far the most effective reagents to induce antigen specific anergy in murine and human systems are CTLA4Ig and Fab anti-CD28, whereas anti-B7 monoclonal - antibodies have been much less effective (Harding, F.A., et al. (1992) Nature. ~, 607-609;
Lenschow, D.J., et al. (1992) Science. ~1, 789-792, Chen, L., et al. (1992) Cell. 71, 1093-1102; Tan, P., et al. (1993) J. Exp. Med. 177, 165-173.). These observations are also consistent with the hypothesis that alternative CTLA4/CD28 ligands capable of inducing IL-2 exist, and taken together with the results p.~sellled herein, suggest that all three CTLA4 binding counter-receptors may be critical for the induction of T cell immllnity. Furthermore, 21~7~ 70 their blockade will likely be required for the induction of T cell anergy. Identical results have been observed in the murine system with the identification of two CTLA4 binding lig~n~lc, corresponding to the human B7-1 and B7-2 molecules. APCs in the B7 deficient mouse bind to the CTLA4 and can induce IL-2 secretion. Taken together, these observations indicate that multiple CTLA-4 binding counter-receptors exist and sequentially costimulate T
cell activation in the murine system.

Cloni~ Sequenrin~ and F~pression of the R7-2 An~i~en A. Col ~truction of cDN~ T ibrarv A cDNA library was constructed in the pCDM8 vector (Seed, Nature, 329:840 (1987)) using poly (A)+ RNA from the human anti-IgM activated B cells as described (Aruffo et al, Proc. Natl. Acad. Sci. USA, 84:3365 (1987)). Splenic B cells were cultured at 2X106 cells/ml in complete culture media, {RPMI 1640 with 10% heat inactivated fetal calf serum (FCS), 2mM gll-t~mine7 1 mM sodium pyruvate, penicillin (100 units/ml), streptomycin sulfate ( l OO,ug/ml) and ge~ ycin sulfate (5,ug/ml) }, i~ tissue culture flasks and were activated by cro~linking of sIg with affinity purified rabbit anti-human IgM
coupled to Affi-Gel 702 beads (Bio-Rad), Richmond, CA) (Boyd, A.W., et al., (1985) J.
Immunol. 134,1516). Activated B cells were harvested after 1/6, 1/2, 4, 8 12, 24, 48, 72 and 96 hours.
RNA was prepared by homogenizing activated B cells in a solution of 4M guanidinethiocyanate, 0.5% sarkosyl, 25mM EDTA, pH 7.5, 0.13% Sigma anti-foam A, and 0.7%mercaptoethanol. RNA was purified from the homogenate by centrifugation for 24 hour at 32,000 rpm through a solution of 5.7M CsCl, lOmM EDTA, 25mM Na acetate, pH 7. The pellet of RNA was dissolved in 5% sarkosyl, lmM EDTA, lOmM Tris, pH 7.5 and extracted with two volumes of 50% phenol, 49% chloroform, 1% isoamyl alcohol. RNA was ethanol precipitated twice. Poly (A)+ RNA used in cDNA library construction was purified by two cycles of oligo (dT)-cellulose selection.
Complement~ry DNA was synthesized from 5.5,ug of anti-IgM activated human B
cell poly(A)+ RNA in a reaction cont~ining 50mM Tris, pH 8.3, 75mM KCl, 3mM MgC12, lOmM dithiothreitol, 500,uM dATP, dCTP, dGTP, dTTP, 50,ug/ml oligo(dT)12 18, 180units/ml RNasin, and 10,000 units/ml Moloney-MLV reverse transcriptase in a total volume of 55,u1 at 37 for 1 hr. Following reverse transcription, the cDNA was converted to double-stranded DNA by adjusting the solution to 25mM Tris, pH 8.3, lOOmM KCl, SmM MgCl2, 250~M each dATP, dCTP, dGTP, dTTP, 5mM dithiothreitol, 250 units/ml DNA polymerase I, 8.5 units/ml ribonuclease H and incubating at 16 for 2 hr. EDTA was added to 18mM and ~1670~1 the solution was extracted with an equal volume of 50% phenol, 49% chloroform, 1 %
isoamyl alcohol. DNA was precipitated with two volumes of ethanol in the presence of 2.5M
ammonium acetate and with 4 micrograms of linear polyacrylamide as carrier. In addition, cDNA was synthesi7~d from 4~Lg of anti-IgM activated human B cell poly(A)+ RNA in a - 5 reaction cont~ining 50mM Tris, pH 8.8, 50,ug/ml oligo(dT)12 18, 327 units/ml RNasin, and 952 units/ml AMV reverse transcriptase in a total volume of lOO,ul at 42 for 0.67 hr.
Following reverse transcription, the reverse transcriptase was inactivated by heating at 70 for 10 min. The cDNA was converted to double-stranded DNA by adding 320,u1 H20 and 80~11 of a solution of 0. lM Tris, pH 7.5, 25mM MgC12, 0.5M KCl, 250,ug/ml bovine serum albumin, and 50mM dithiothreitol, and adjusting the solution to 200~M each dATP, dCTP, dGTP, dTTP, 50 units/ml DNA polymerase I, 8 units/ml ribonuclease H and incubating at 16, C for 2 hours. EDTA was added to 18 mM and the solution was extracted with an equal volume of 50 % phenol, 49 % chloroform, 1 % isoamyl alcohol. DNA was precipitated with two volumes of ethanol in the presence of 2.5M ammonium acetate and with 4 micrograms of linear polyacrylamide as carrier.
The DNA from 4,ug of AMV reverse transcription and 2,ug of Moloney MLV reverse transcription was combined. Non-selfcomplement~ry BstXI adaptors were added to the DNA
as follows: The double-stranded cDNA from 6,ug of poly(A)+ RNA was incubated with 3.6,u g of a kin~ee~l oligonucleotide ofthe sequence CTTTAGAGCACA (SEQ ID NO:15) and 2.4 ~Lg of a kin~ce~l oligonucleotide of the sequence CTCTAAAG (SEQ ID NO: 16) in a solution cont~ining 6mM Tris, pH 7.5, 6mM MgC12, SmM NaCl, 35011g/ml bovine serum albumin, 7mM mercaptoethanol, O.lmM ATP, 2mM dithiothreitol, lmM spermicline~ and 600 units T4 DNA ligase in a total volume of 0.45ml at 15 C for 16 hours. EDTA was added to 34mM
and the solution was extracted with an equal volume of 50% phenol, 49% chloroform, 1 %
isoamyl alcohol. DNA was precipitated with two volumes of ethanol in the presence of 2.5M
ammonium acetate.
DNA larger than 600bp was selected as follows: The adaptored DNA was redissolvedin lOmM Tris, pH 8, lmM EDTA, 600mM NaCl, 0.1% sarkosyl and chromatographed on aSepharose CL-4B column in the same buffer. DNA in the void volume of the column (cont~ining DNA greater than 600bp) was pooled and ethanol precipitated.
The pCDM8 vector was prepared for cDNA cloning by digestion with BstXI and purification on an agarose gel. Adaptored DNA from 6~1g of poly(A)+RNA was ligated to 2.25~1g of BstXI cut pCDM8 in a solution cont~ining 6mM Tris, pH 7.5, 6mM MgC12, SmM
NaCl, 350~Lg/ml bovine serum albumin, 7mM mercaptoethanol, O.lmM ATP, 2mM
dithiothreitol, lmM spermidine, and 600 units T4 DNA ligase in a total volume of 1.5ml at 15 for 24 hr. The ligation reaction mixture was transformed into competent E.coli MC 1061/P3 and a total of 4,290,000 independent cDNA clones were obtained.

WO 9s/03408 2~ 1 PCTIUS94/08423 Plasmid DNA was prepared from a 500 ml culture of the original transformation ofthe cDNA library. Plasmid DNA was purified by the ~lk~line Iysis procedure followed by twice banding in CsCl equilibrium gradients (Maniatis et al, Molecular Cloning: A
Laboratory Manual, Cold Spring Harbor, NY (1987)).
S
P~. Clonin~ Procedure In the first round of screening, thirty 100 mm dishes of 50% confluent COS cells were transfected with O.O5~1g/ml anti-IgM activated human B cells library DNA using the DEAE-Dextran method (Seed et al, Proc. Nafl. Acad. Sci USA, 84:3365 (1987)). The cells were 10 Lly~.si~ l and re-plated after 24 hours. After 47 hours, the cells were ~let~t~.h~d by in~ub~tion in PBS/0.5 mM EDTA, pH 7.4/0.02% Na azide at 37C for 30 min. The let~ched cells were treated with 10 ~g/ml/CTLA4Ig and CD28Ig for 45 minntes at 4C. Cells were washed and distributed into panning dishes coated with affinity-purified Goat anti-human IgG antibody and allowed to attach at room le~ e~ lt;. After 3 hours, the plates were gently washed twice with PBS/O.SmM EDTA, pH 7.4/0.02% Na azide, 5% FCS and once with O.l5M NaCl, 0.01 M Hepes, pH 7.4, 5% FCS. Episomal DNA was recovered from the panned cells and transformed into E. coli DHlOB/P3. The plasmid DNA was re-introduced into COS cells via spheroplast fusion as described (Seed et al, Proc. Natl. Acad. Sci. USA, 84:3365 (1987)) and the cycle of ~A~l~s~ion and panning was repeated twice. In the second and third rounds of selection, after 47 hours, the det~hecl COS cells were first incubated with a-B7-1 mAbs (133 and Bl.1, 10 ~g/ml~, and COS cells ~A~ S~illg B7-1 were removed by a-mouse IgG and IgM coated magnetic beads. COS cells were then treated with 10 ,ug/ml of human CTLA4Ig (hCTLA4Ig) and human CD28Ig (hCD28Ig) and hurnan B7-2 ~A~les~ing COS cells were selected by p~nning on dishes with goat anti-human IgG antibody plates.
After the third round, plasmid DNA was ple~aled from individual colonies and kansfected into COS cells by the DEAE-Dexkan method. Expression of B7-2 on transfected COS cells was analyzed by indirect immunt~fluorescence with CTLA4Ig.
After the final round of selection, plasmid DNA was prepared from individual colonies. A total of 4 of 48 ç~n~ te clones contained a cDNA insert of approximately 1.2 kb. Plasmid DNA from these four clones was kansfected into COS cells. All four clones were skongly positive for B7-2 expression by indirect immunofluorescence using CTLA4Ig and flow cytometric analysis.

C. Sequencin~
The B7-2 cDNA insert in clone29 was sequenced in the pCDM8 expression vector employing the following skategy. Initial sequencing was performed using sequencing primers T7, CDM8R (Invikogen) homologous to pCDM8 vector sequences adjacent to the ' 2~6~0~

cloned B7-2 cDNA (see Table I). Sequencing was performed using dye terminator chemistr~
and an ABI automated DNA sequencer. (ABI, Foster City, CA). DNA sequence obtained using these primers was used to design additional sequencing primers (see Table I). This cycle of sequencing and selection of additional primers was continued until the B7-2 cDNA
5 was completely sequenced on both strands.

TABLE I

T7(F) (SE~Q ID NO:3) S'drTAATACGACTCACTATAGGG]3' 10 CDM8(R) (SEQ ID NO:4) 5'd[TAAGGTTCCTTCACAAAG]3' CDM8 RGV(2) (SEQ ID NO:5) S'd[ACTGGTAGGTATGGAAGATCC]3' HBX29-5P (2R) (SEQ ID NO:6) 5'd~ATGCGAATCATTCCTGTGGGC]3' HBX29-5P (2F) (SEQ ID NO:7) 5'd[AAAGCCCACAGGAATGATTCG]3' HBX29-5P (SEQ ID NO:8) 5'd[CTCTCAAAACCAAABCCTGAG]3' SPA (SEQ ID NO:9) 5'd[TTAGGTCACAGCAGAAGCAGC]3' 5PA (3FA) (SEQ ID NO:10) 5'd[TCTGGAAACTGACAAGACGCG~3' HBX29-SP(lR) (SEQ ID NO:11) 5'd[CTCAGGCTTTGGTTTTGAGAG]3' HBX29-3P(lR) (SEQ ID NO: 12) 5'd[CACTCTCTTCCCTCTCCATTG]3' HBX29-5P(3R) (SEQ ID NO:13) 5'd[GACAAGCTGATGGAAACGTCG]3' HBX29-3P(lP) (SEQ ID NO:14) 5'd~CAATGGAGAGGGAAGAGAGTG]3' The human B7-2 clone 29 contained an insert of 1,120 base pairs with a single long open reading frame of 987 nucleotides and approximately 27 nucleotides of 3' noncoding sequences (Figure 8 (SEQ ID NO: 1)). The predicted amino acid sequence encoded by the open reading frarne of the protein is shown below the nucleotide sequence in Figure 8. The encoded protein, human B7-2, is predicted to be 329 amino acids in length (SEQ ID NO:2).
This protein sequence exhibits many features common to other type 1 Ig ~u~ r~llily embrane proteins. Protein translation is predicted to begin at the ATG codon (nucleotide 107-109) based on DNA homology in this region with the consensus eukaryotic translation initiation site (Kozalc, M. (1987) ~ucl. Acids Res. 1~:8125-8148)- The amino terminus of the human B7-2 protein (amino acids 1 to 23) has the characteristics of a secretory signal peptide with a predicted cleavage between the ~l~nines at positions 23 and 24 (von Heiine (1986) Nucl. Acids Res. 14:4683). Processing at this site would result in a human B7-2 membrane bound protein of 306 amino acid with an unmodifed molecular weight of approximately 34 kDa. This protein would consist of an extracellular Ig superfamily V and C like domains, of from about amino acid residue 24^245, a hydrophobic tr~n.~membrane domain of from about wo 95,03408 2 ~ PCT/US94/08423 amino acid residue 246-268 and a long cytoplasmic domain of from about amino acid residue 269-329. The homologies to the Ig superfamily are due to the two contiguous Ig-like domains in the extracellular region bound by the cysteines at positions 40 to 110 and 157 to 218. The extracellular domain also contains eight potential N-linked glycosylation sites. E.
5 coli transfected with a vector cont~inin~ the cDNA insert of clone 29, encoding the human B7-2 protein, was deposited with the American Type Culture Collection (ATCC) on July 26, 1993 as Accession No. 69357.
Comparison of both the nucleotide and amino acid sequences of human B7-2 with the GenBank and EMBL databases showed that only the human and murine B7-1 proteins are related. Ali~nment of the three B7 protein sequences (see Figure 13) shows that human B7-2 has approximately 26% amino acid identity with human B7-1. Figure 13 represents the comparison of the amino acid sequences for human B7-2 (hB7-2) (SEQ ID NO:2), human B7-1 (hB7-1) (SEQ ID NO: 28 and 29) and murine B7 (mB7) (SEQ ID NO: 30 and 31). The amino acid sequences for the human B7-1 and murine B7 (referred to herein as murine B7-1) can be found in Genbank at Accession #M27533 and X60958 respectively. Vertical lines in Figure 13 show iclentir~l amino acids between the hB7-2 and hB7-1 or mB7. Identical amino acids between hB7-1 and mB7 are not shown. The hB7-2 protein exhibits the same general structure as hB7-1 as defined by the common cysteines (positions 40 and 110, IgV domains;
positions 157 and 217, IgC domain) which the Ig ~u~elr~llily domains and by many other 20 common amino acids. Since both hB7-1 and mB7 have been shown to bind to both human CTLA4 and human CD28, the amino acids in common between these two related proteins will be those necessary to comprise a CTLA4 or CD28 binding sequence. An example of such a sequence would be the KYMGRTSFD (position 81-89, hB7-2) (SEQ ID NO:17) orKSQDNVTELYDVS (position 188-200, hB7-2) (SEQ ID NO: 18). Additional related 25 sequences are evident from the sequence comparison and others can be inferred by con~ rin~ homologous related amino acids such as aspartic acid and glutamic acid, alanine and glycine and other recognized functionally related amino acids. The B7 sequences share a highly positive charged domain with the cytoplasmic portion WKWKKKKRPRNSYKC
(position 269-282, hB7-2) (SEQ ID NO:19) which is probably involved in intracellular 30 ~i~n~lin~

Ch~racteri~ on of the Recombin~nt B7-2 Ant~en ~ B7-~ R;nds CT~ ~4I~ and Not Anti-R7-1 and Anti-n7-3 Monoclonal Antibodies COS cells transfected with either vector DNA (pCDNAI), or an expression plasmid cont~inin~ B7-1 (B7-1) or B7-2 (B7-2) were prepared. After 72 hours, the transfected COS

WO 95/03408 ~ ~ 6 7 0 9 1 PCT/US94/08423 cells were detached by incubation in PBS cont~ining 0.5 mM EDTA and 0.02% Na azide for 30 min. at 37C. Cells were analyzed for cell surface e~-~ssion by indirect imml-nofluorescence and flow cytometric analysis using fluoroscein isothiocyanate conjugated (FITC) goat-anti-mouse Ig or goat-anti-human IgG FITC (Figure 9). Cell surface - 5 t;~yLc;ssion of B7-1 was detected with mAbs 133 (anti-B7-1) and BB-1 (anti-B7-1 and anti-B7-3) and with CTLA4Ig, whereas B7-2 reacted only with CTLA4Ig. Neither of the B7 transfectants showed any staining with the isotype controls (IgM or control Ig). The vector transfected COS cells showed no st~ining with any of the detection reagents. In addition, none of the cells showed any staining with the FITC labeled detection reagents and alone.
This demonstrates that B7-2 encodes a protein that is a CTLA4 counter-receptor but is distinct from B7-1 and B7-3.

R. RNA Rlot ~n~lysiS of P~7-2 Fxpre~ion in Un~timulated ~nd Activated Hl-m~n R Cells~
C~ell T ines ~ntl Myelom~
Human splenic B cells were isolated by removing T cells and monocytes as previously described (Free~lm~n, A.S., Freeman, G.J., Horowitz, J.C., Daley, J., Nadler, L.M., J. Immunol. (1987) 1;~:3260-3267). Splenic B cells were activated using anti-Ig beads and cells were harvested at the indicated times (Free-lm~n et al., (1987), cited supra). Human myelomas from bone marrow specimens were enriched by removing T cells and monocytes using E rosettes and adherence as previously described (Freeman, G.J., et al., J. Immunol.
(1989) 143:2714-2722). RNA was prepared by gl~ni(1ine thiocyanate homogenization and cesium chloride centrifugation. Equal amounts of RNA (2011g) were electrophoresed on an agarose gel, blotted, and hybridized to 32P-labelled B7-2 cDNA. Figure 10, panel a, shows RNA blot analysis of Im~timl7l~ted and anti-Ig activated human splenic B cells and of cell lines including Raji (B cell Burkitts lymphoma), Daudi (B cell Burkitt's lymphoma), RPMI
8226 (myeloma), K562 (erythrole-lk~mi~), and REX (T cell acute lymphoblastic leukemia).
Figure 10, panel b shows RNA blot analysis of human myeloma specimens.
Three mRNA transcripts of 1.35, 1.65 and 3.0 kb were identified by hybridization to the B7-2 cDNA (Figure 10, panel b). RNA blot analysis demonstrated that B7-2 mRNA is expressed in lln~tim~ ted human splenic B cells and increases 4-fold following activation (Figure 10, panel a). B7-2 mRNA was expressed in B cell neoplastic lines (Raji, Daudi) and a myeloma (RPMI 8226) but not in the erythroleukemia K562 and the T cell line REX. In contrast, we have previously shown that B7-1 mRNA is not expressed in resting B cells and is transiently expressed following activation (G.J. Freeman et al. (1989) supra). F~min~tion of mRNA isolated from human myelomas demonstrates that B7-2 mRNA is expressed in 6 of 6 p~ti~nt~, whereas B7-1 was found in only 1 ofthese 6 (G.J. Freeman et al. (1989) supra).
Thus, B7-1 and B7-2 e~ s~ion appears to be independently regulated.

Wo 95/03408 PCT/US94/08423 C. Costimulation Human CD28+ T cells were isolated by immunomagnetic bead depletion using monoclonal antibodies directed against B cells, natural killer cells and macrophages as previously described (Gimmi, C.D., et al. (1993) Proc. Natl. Acad. Sci. USA ~, 6586-6590).
B7-1, B7-2 and vector transfected COS cells were harvested 72 hours after transfection, incubated with 2511g/ml of mitomycin-C for 1 hour, and then extensively washed. 105 CD28+ and T cells were incubated with 1 ng/ml of phorbol myristic acid (PMA) and the in~lic~ted number of COS transfectants (Figure 11). As shown in Figure 11, panel a, T cell proliferation was measured by 3H-thymidine (1 ~Ci) incorporated for the last 12 hours of a 72 hour incubation. Panel b of Figure 11 shows IL-2 production by T cells as measured by ELISA (Biosource, CA) using supernatants harvested 24 hours after the initiation of culture.

n. P,7-2 Costimulation i~ n~t Blocked by Anti-B7-1 and Anti-P~7-3 mAbs but is Blocked by CTT ~4-I~ and Anti-CD28 Fab Human CD28+ T cells were isolated by immunomagnetic bead depletion using mAbs directed against B cells, natural killer cells, and macrophages as previously described (Gimmi, C.D., Freeman, G.J., Gribben, J.G., Gray, G., Nadler, L.M. (1993) Proc. Natl. Acad.
Sci USA 90, 6586-6590). B7-1, B7-2, and vector transfected COS cells were harvested 72 hours after transfection, incubated with 25,ug/ml of mitomycin-C for I hour, and then extensively washed. 105 CD28+ T cells were incubated with 1 ng/ml of phorbol myristic acetate (PMA) and 2 x 104 COS transfectants. Blocking agents (lO~g/ml) are indicated on the left side of Figure 12 and include: 1) no monoclonal antibody (no blocking agents), 2) mAb 133 (anti-B7-1 mAb), 3) rnAb BBl (anti-B7-1 and anti-B7-3 mAb), 4) mAb B5 (control IgM mAb), 5) anti-CD28 Fab (mAb 9.3), 6) CTLA-Ig, and 7) control Ig. Panel a of Figure 12 shows proliferation measured by 3H-thymidine (l,uCi) incorporation for the last 12 hours of a 72 hour incubation. Figure 12, panel b, shows IL-2 production æ measured by ELISA (Biosource~ CA) using supern~t~nt~ harvested 24 hours after the initiation of culture.
B7-1 and B7-2 transfected COS cells costim~ te~l equivalent levels of T cell proliferation when tested at various stimulator to responder ratios (Figure 11). Like B7-1, B7-2 transfected COS cell costimulation resulted in the production of IL-2 over a wide range of stimul~tor to responder cell ratios (Figure 11). In contrast, vector transfected COS cells did not costimulate T cell proliferation or IL-2 production.

Wo 95/03408 ~16 7 ~ 91 PCT/US94/08423 F. P~7-2 Costimulation is not Blocked by Anti-B7-1 and Anti-~7-3 mAbs but is Blocked by CTT ~4-I~ ~nd ~nti-CD28 Fab Human CD28+ T cells were isolated by immunomagnetic bead depletion using mAbs directed against B cells, natural killer cells, and macrophages as previously described - 5 (Gimmi, C.D., Freeman, G.J., Gribben, J.G., Gray, G., Nadler, L.M. (1993) Proc. Natl. Acad Sci USA ~, 6586-6590). B7-1, B7-2, and vector transfected COS cells were harvested 72 hours after transfection, incubated with 25~1g/ml of mitomycin-C for 1 hour, and then extensively washed. 105 CD28+ T cells were incubated with 1 ng/ml of phorbol myristic acetate (PMA) and 2 x 104 COS transfectants. Blocking agents (lOIlg/ml) are indicated on the left side of Figure 12 and include: 1) no monoclonal antibody (no blocking agents), 2) mAb 133 (anti-B7-1 mAb), 3) mAb BB1 (anti-B7-1 and anti-B7-3 mAb), 4) mAb B5 (control IgM mAb), 5) anti-CD28 Fab (mAb 9.3), 6) CTLA-Ig, and 7) control Ig. Panel a of Figure 12 shows proliferation measured by 3H-thymidine (l~Ci) incorporation for the last 12 hours of a 72 hour incubation. Figure 12, panel b, shows IL-2 production as measured by ELISA (Biosource, CA) using sUpern~t~nt~ harvested 24 hours after the initiation of culture.
To distinguish B7-2 from B7-1 and B7-3, mAbs directed against B7-1 and B7-3 wereused to inhibit proliferation and IL-2 production of submitogenically activated human CD28+
T cells. Both B7-1 and B7-2 COS tran~r~ck~ costimulated T cell proliferation and IL-2 production (Figure 12). MAbs 133 (Free~lm~n, A.S. et al. (1987) supra) (anti-B7-1) and BBl (Boussiotis, V.A., et al., (in review) Proc. Natl. Acad. Sci. USA; Yokochi, T., Holly, R.D., Clark, E.A. (1982) J. Immunol. 128, 823-827) (anti-B7-1 and anti-B7-3) completely inhibited proliferation and IL-2 secretion induced by B7-1 but had no effect upon costim~ ion by B7-2 transfected COS cells. Isotype m~trhP~l control B5 mAb had no effect. To ~letermine whether B7-2 signals via the CD28/CTLA4 pathway, anti-CD28 Fab and CTLA4-Ig fusion protein were tested to cl~t~rmine whether they inhibited B7-2 costimlll~tion. Both anti-CD28 Fab and CTLA4-Ig inhibited proliferation and IL-2 production intlll~e~l by either B7-1 or B7-2 COS transfectants whereas control Ig fusion protein had no effect (Figure 12). While CTLA4-Ig inhibited B7-2 costimlll~tion of proliferation by only 90%, in other experiments inhibition was more pronounced (98-100%). None of the blocking agents inhibited T cell proliferation or IL-2 production induced by the combination of PMA and phytohem~g~lutinin.
Like B7-1, B7-2 is a counter-receptor for the CD28 and CTLA4 T cell surface molecules. Both proteins are similar in that they are: 1) expressed on the surface of APCs;
2) structurally related to the Ig supergene family with an IgV and IgC domain which share 26% arnino acid identity, and 3) capable of costimulating T cells to produce IL-2 and proliferate. However, B7-1 and B7-2 differ in several fimc1~ment~l ways. First, B7-2 mRNA
is co~ iLulively expressed in unstimulated B cells, whereas B7-1 mRNA does not appear 21~9 1 -78-until 4 hours and cell surface protein is not detected until 24 hours (Free~1m~n? A.S., et al.
(1987) supra; Freeman, G.J., et al. (1989) supra). Unstim~ te~l hurnan B cells do not express CTLA4 counter-receptors on the cell surface and do not costim~ te T cell proliferation (Boussiotis, V.A., et al. supra). Therefore, expression of B7-2 mRNA in unstimulated B cells would allow rapid expression of B7-2 protein on the cell surface following activation, presurnably from stored mRNA or protein. Costimlll~tion by B7-2 transfectants is partially sensitive to paraformaldehyde fixation, whereas B7-2 costimlll~tion is resistant (Gimmi, C.D., et al. (1991) Proc. Natl. Acad. Sci USA 88, 6575-6579). Second, ~xl~les~ion of B7-1 and B7-2 in cell lines and human B cell neoplasms substantially differs. Third, B7-2 protein contains a longer cytoplasmic domain than B7-1 and this could play a role in ~i~n~lin~ B-cell di~,ellliation. These phenotypic and functional differences suggest that these homologous molecules may have biologically distinct functions.

Clor~i~ and Sequenl~in~ of the Murine B7-2 Al~tu~en A. Construction of cDNA T ibr~ly A cDNA library was constructed in the pCDM8 vector (Seed, Nature, 329:840 (1987)) using poly (A)+ RNA from dibutryl cyclic AMP (cAMP) activated M12 cells (a murine B cell tumor line) as described (Aruffo et al, Proc. Natl. Acad. Sci USA, 84:3365 (1987)).
M12 cells were cultured at 1x106 cells/ml in complete culture media, {RPMI 1640 with 10% heat inactivated fetal calf serum (FCS), 2mM glllt~mine, 1 mM sodium pyruvate, penicillin (100 unitslml), ~Llc;~Loll~ycin sulfate (lOO~g/ml) and gen~..ycin sulfate (5,ug/ml)3, 25 in tissue culture flasks and were activated by 300~1g/ml dibutryl cAMP (Nabavi, N., et al.
(1992) Nature ~Q., 266-268). Activated M12 cells were harvested after 0, 6, 12, 18, 24 and 30 hours.
RNA was prepared by homogenizing activated M12 cells in a solution of 4M
gll~niclin~ thiocyanate, 0.5% sarkosyl, 25mM EDTA, pH 7.5, 0.13% Sigma anti-foam A, and 30 0.7% mercaptoethanol. RNA was purified from the homogenate by centrifugation for 24 hour at 32,000 rpm through a solution of 5.7M CsCI, lOmM EDTA, 25mM Na acetate, pH 7.
The pellet of RNA was dissolved in 5% sarkosyl, lmM EDTA, lOmM Tris, pH 7.5 and extracted with two volumes of 50% phenol, 49% chloroform, 1% isoamyl alcohol. RNA was ethanol precipitated twice. Poly (A)+ RNA used in cDNA library construction was purified 35 by two cycles of oligo (dT)-cellulose selection Complement~ry DNA was synthesi7ecl from S.S,ug of dibutryl cAMP activated murine M12 cell poly(A)+ RNA in a reaction co"~ i"~ SOmM Tris, pH 8.3, 75mM KCl, WO 95/0340~ 2 ~ 6 7 ~ 91 PCT/US94tO8423 .~

3mM MgC12, 1 OmM dithiothreitol, SOO~lM dATP, dCTP, dGTP, dTTP, 50~g/ml oligo(dT)12 18, 180 units/ml RNasin, and 10,000 units/ml Moloney-MLV reverse transcriptase in a total volume of SS~ll at 37 C for 1 hr. Following reverse transcription, the cDNA was converted to double-stranded DNA by adjusting the solution to 25mM Tris, pH
8.3, l OOmM KCl, SmM MgC12, 250~1M each dATP, dCTP, dGTP, dTTP, SmM
dithiothreitol, 250 units/ml DNA polymerase I, 8.5 units/ml ribonuclease H and incubating at 16 C for 2 hr. EDTA was added to 18mM and the solution was extracted with an equal volume of 50% phenol, 49% chloroform, 1 % isoamyl alcohol. DNA was precipitated with two volurnes of ethanol in the presence of 2.5M ammonium acetate and with 4 micrograms of linear polyacrylamide as carrier. Following reverse transcription, the reverse transcriptase was inactivated by heating at 70 C for 10 min. The cDNA was converted to double-stranded DNA by adding 320~L1 H20 and 80,u1 of a solution of O.lM Tris, pH 7.5, 25mM MgC12, O.SM KCI, 250~1g/ml bovine serum albumin, and 50mM dithiothreitol, and adjusting the solution to 200~M each dATP, dCTP, dGTP, dTTP, SO units/ml DNA polymerase I, 8 units/ml ribonuclease H and incllbating at 16 C for 2 hours. EDTA was added to 18 mM and the solution was extracted with an equal volume of 50% phenol, 49% chloroform, 1%
isoamyl alcohol. DNA was precipitated with two volumes of ethanol in the presence of 2.5M
ammonium acetate and with 4 micrograms of linear polyacrylamide as carrier.
2~Lg of non-selfcomplçment~ry BstXI adaptors were added to the DNA as follows:
The double-stranded cDNA from 5.5~1g of poly(A)+ RNA was incubated with 3.6,~Lg of a kin~t~ecl oligonucleotide of the sequence Cl-l~AGAGCACA (SEQ ID NO: 15) and 2.4~g of a kin~eed oligonucleotide ofthe sequence CTCTAAAG (SEQ ID NO:16) in a solution co..l~t;~ 6mM Tris, pH 7.5, 6mM MgC12, 5mM NaCl, 350,ug/ml bovine serum albumin, 7mM mercaptoethanol, O.lmM ATP, 2mM dithiothreitol, lmM spermi(line, and 600 units T4 DNA ligase in a total volume of 0.45ml at 15 for 16 hours. EDTA was added to 34mM and the solution was extracted with an equal volume of 50% phenol, 49% chloroform, 1%
isoamyl alcohol. DNA was precipitated with two volumes of ethanol in the presence of 2.5M
ammoniD acetate.
DNA larger than 600bp was selected as follows: The adaptored DNA was redissolvedin lOmM Tris, pH 8, lmM EDTA, 600mM NaCl, 0.1% sarkosyl and chromatographed on aSepharose CL-4B column in the same buffer. DNA in the void volume of the column (co--l~i.,;t~f~ DNA greater than 600bp) was pooled and ethanol precipitated.
The pCDM8 vector was prepared for cDNA cloning by digestion with BstXI and purification on an agarose gel. Adaptored DNA from 5.5,ug of poly(A)+RNA was ligated to 2.25~1g of BstXI cut pCDM8 in a solution co~ 6mM Tris, p~J 7.5, 6mM MgC12, SmM
NaCl, 350~Lg/ml bovine serum albumin, 7mM mercaptoethanol, O.lmM ATP, 2mM
dithiothreitol, lmM sperrni~line, and 600 units T4 DNA ligase in a total volume of 1.5ml at wo 95,03408 ~ ~ 6 7 ~ ~ ~ PCT/US94/08423 ~

15 for 24 hr. The ligation reaction mixture was transformed into competent E.coli MC1061/P3 and a total of 200 x 106 independent cDNA clones were obtained.
Plasmid DNA was prepared from a 500 ml culture of the original transformation ofthe cDNA library. Plasmid DNA was purified by the ~lk~line Iysis procedure followed by 5 twice banding in CsCl equilibrium gradients (Maniatis et al, Molecular Cloning: A
Laboratory Manual, Cold Spring Harbor, NY (1987)).

R. Clonin~ Proce~lllre In the first round of screening, thirty 100 mm dishes of 50% confluent COS cells were transfected with 0.05~1g/ml activated M12 murine B cell library DNA using the DEAE-Dextran method (Seed et al, Proc. Natl. Acad. Sci. US~, 84:3365 (1987)). The cells were trypsini7~cl and re-plated after 24 hours. After 47 hours, the cells were detached by incubation in PBS/0.5 mM EDTA, pH 7.4/0.02% Na azide at 37C for 30 min. The detached cells were treated with 10 ~g/ml/human CTLA4Ig and murine CD28Ig for 45 minutes at 4C.
Cells were washed and distributed into p~nning dishes coated with affinity-purified Goat anti-human IgG antibody and allowed to attach at room temperature. After 3 hours, the plates were gently washed twice with PBS/O.SmM EDTA, pH 7.4/0.02% Na azide, 5% FCS and once with 0.15M NaCl, 0.01 M Hepes, pH 7.4, 5% FCS. Episomal DNA was recovered from the panned cells and transformed into E. coli DHlOB/P3. The plasmid DNA was re-introduced into COS cells via spheroplast fusion as described (Seed et al, Proc. Natl. Acad.
Sci USA, 84:3365 (1987)) and the cycle of expression and p~nnin~ was repeated twice. In the second and third rounds of selection, after 47 hours, the detached COS cells were first incubated with a-murine B7-1 mAb (16-lOA1, 10 ,ug/ml), and COS cells expressing B7-1 were removed by a-mouse IgG and IgM coated magnetic beads. COS cells were then treated with lO~Lg/ml of human CTLA4Ig and murine CD28Ig and murine B7-2 expressing COS
cells were selected by p~nning on dishes coated with goat anti-human IgG antibody. After the third round, plasmid DNA was prepared from individual colonies and transfected into COS cells by the DEAE-Dextran method. Expression of B7-2 on transfected COS cells was analyzed by indirect immllnnfluorescence with CTLA4Ig.
After the final round of selection, plasmid DNA was prepared from individual colonies. A total of 6 of 8 candidate clones contained a cDNA insert of approximately 1.2 kb. Plasmid DNA from these eight clones was transfected into COS cells. All six clones with the 1.2 Kb cDNA insert were strongly positive for B7-2 ~iession by indirectimmunofluorescence using CTLA4Ig and flow cytometric analysis.

9 ~
~YO 95/03408 PCT/US94/08423 C. Se~uencin~
The B7-2 cDNA insert in clone4 was sequenced in the pCDM8 expression vector employing the following strategy. Initial sequencing was performed using sequencing primers T7, CDM8R (Invitrogen) homologous to pCDM8 vector sequences adjacent to the 5 cloned B7-2 cDNA (see Table II). Sequencing was performed using dye termin~torchPmi~try and an ABI automated DNA sequencer. (ABI, Foster City, CA). DNA sequence obtained using these primers was used to design additional sequencing primers (see Table II).
This cycle of sequencing and selection of additional primers was continl1c-1 until the murine B7-2 cDNA was completely sequenced on both strands.
TABLE II

T7(F) (SEQ ID NO:3) 5'd[TAATACGACTCACTATAGGG]3' CDM8(R) (SEQ ID NO:4) 5'd[TAAGGTTCCTTCACAAAG]3' MBX4-lF (SEQ ID NO:24) 5'd[ACATAAGCCTGAGTGAGCTGG]3' MBX4-2R (SEQ ID NO:25) 5'd[ATGATGAGCAGCATCACAAGG]3' MBX4-14 (SEQ ID NO:26) 5'd[TGGTCGAGTGAGTCCGAATAC]3' MBX4-2F (SEQ ID NO:27) 5'd[GACGAGTAGTAACATACAGTG]3' A murine B7-2 clone (mB7-2, clone 4) was obtained cont~inin~ an insert of 1,163 base pairs with a single long open reading frame of 927 nucleotides and approximately 126 nucleotides of 3' noncoding sequences (Figure 14, SEQ ID NO:22). The predicted amino acid sequence encoded by the open reading frame of the protein is shown below the nucleotide sequence in Figure 14. The encoded murine B7-2 protein, is predicted to be 309 amino acid residues in length (SEQ ID NO:23). This protein sequence exhibits many features common to other type I Ig superfamily membrane proteins. Protein translation is predicted to begin at the methionine codon (ATG, nucleotides 111 to 113) based on the DNA
homology in this region with the consensus eucaryotic translation initiation site (see Kozak, M. (1987) Nucl. Acids Res. 15:8125-8148). The amino terminus ofthe murine B7-2 protein (amino acids 1 to 23) has the characteristics of a secretory signal peptide with a predicted cleavage between the alanine at position 23 and the valine at position 24 (von Heijne (1987) Nucl. ~cids Res. 14:4683). Processing at this site would result in a murine B7-2 membrane bound protein of 286 amino acids having an unmodified molecular weight of approximately - 32 kDa. This protein would consist of an approximate extracellular Ig superfamily V and C
like domains of from about amino acid residue 24 to 246, a hydrophobic transmembrane domain of from about amino acid residue 247 to 265, and a long cytoplasmic domain of from about amino acid residue 266 to 309. The homologies to the Ig superfamily are due to the WO 95/03408 ~ PCT/US94/08423 two contiguous Ig-like domains in the extracellular region bound by the cysteines at positions 40 to 110 and 157 to 216. The extracellular domain also contains nine potential N-linked glycosylation sites and~ like murine B7-1, is probably glycosylated. Glycosylation ofthe murine B7-2 protein may increase the molecular weight to about 50-70 kDa. The cytoplasmic domain of murine B7-2 contains a common region which has a cysteine followed by positively charged amino acids which presumably functions as ~ign~ling or regulatory domain within an APC. Comparison of both the nucleotide and amino acid sequences of mur~ne B7-2 with the GenBank and EMBL ~l~t~b~es yielded significanthomology (about 26% amino acid sequence identity) with human and murine B7-1. Murine B7-2 exhibits about 50% identity and 67% similarity with its human homologue, hB7-2. E.
coli (DH106/p3) transfected with a vector (plasmid pmBx4) con~ining a cDNA insert encoding murine B7-2 (clone 4) was deposited with the American Type Culture Collection (ATCC) on August 18, 1993 as Accession No. 69388.

n. Costimulation CD4+ murine T cells were purified by first depleting red blood cells by tre~tment with Tris-NH4Cl. T cells were enriched by passage over a nylon wool column. CD4+ T
cells were purified by two-fold tre~tment with a lllixlu~e of anti-MHC class II and anti-CD28 mAbs and rabbit complement. Murine B7-1 (obtained from Dr. Gordon Freeman, Dana-Farber Cancer Institute, Boston, MA; see also, Freeman, G.J. et al (1991) J. Exp. Med. 174, 625-631) murine B7-2, and vector transfected COS cells were harvested 72 hours after trnasfection, incubated with 25,ug/ml mitomycin-C for one hour, and then extensively washed. 105 murine CD4+ T cells were incubated with lng/ml of phorbol myristic acid (PMA) and 2 x 104 COS transfectants (Table III). T cell proliferation was measured by 3H-thymidine (l~lCi) incorporated for the last 12 hours of a 72 hour incubation.

TABLE III
3H-Thymidine Tncorporation (cpm) CD4+ T cells 175 CD4+ T cells + lng/ml PMA 49 CD4+ T cells + COS-vector 1750 CD4+ T cells + COS-B7-1 4400 CD4+ T cells + COS-B7-2 2236 CD4+ T cells + lng/ml PMA + COS-vector 2354 CD4+ T cells + lng/ml PMA + COS-B7-l67935 CD4+ T cells + lng/ml PMA + COS-B7-243847 O 95/03408 ~ 1 6 7 0 9 ~L PCT/US94/08423 Con~tru~tion and Ch~r~cteri7~tion of H~lman B7-2 Imlnuno~Jobulin Fusion Proteins ,; ..
A. PreparationOfH-Im~nF~7-2IgFusionProtein~
The extracellular portion of human B7-2 was prepared as a fusion protein coupled to an immllnoglobulin constant region. The immnnc)globulin constant region may contain genetic modifications including those which reduce or elimin~te effector activity inherent in the immunoglobulin skucture. Briefly, DNA encoding the extracellular portion of hB7-2 was joined to DNA encoding the hinge, CH2 and CH3 regions of human IgC~l or IgC~4 modified by directed mutagenesis. This was accomplished as described in the following subsections.

B. Preparation of Gene Fusions DNA fragments corresponding to the DNA sequences of interest were prepared by polymerase chain reaction (PCR) using primer pairs described below. In general, PCR
reactions were prepared in 100 1ll final volume composed of Taq, polymerase buffer (Gene Amp PCR Kit, Perkin-Elmer/Cetus, Norwalk, CT) cont~ining primers (1 ,uM each), dNTPs (200 ~lM each) 1 ng oftemplate DNA, and Taq, polymerase (Saiki, R.K., et al. (1988) Science 239:487-491). PCR DNA amplifications were run on a thermocycler (Ericomp, San Diego, CA) for 25 to 30 cycles each composed of a denaturation step (1 minute at 94C), a renaturation step (30 seconds at 54C), and a chain elongation step (1 minute at 72C). The skucture of each hB7-2 Ig genetic fusion consisted of a signal sequence to facilitate secretion coupled to the exkacellular domain of B7-2 and the hinge, CH2 and CH3 domains of human IgC~l or IgCy4. The IgC gamma 1 and IgC gamma 4 sequences contained nucleotide changes within the hinge region to replace cysteine residues available for disulfide bond formation with serine residues and may contain nucleotide changes to replace amino acids within the CH2 domain thought to be required for IgC binding to Fc receptors andcomplement activation.
Sequence analysis confirmed structures of both m~4 and ~1 clones, and each construct was used to kansfect 293 cells to test transient expression. hIgG ELISA measured/confirmed transient ex~)lc~sion levels approximately equal to 100 ng protein/ml cell supernatant for both constructs. NSO cell lines were transfected for permanent ~ sion the the fusion proteins.

- C. Genf~tic Constructioll of h~7-2Ig Fusion Prote;ns (1). Ple~.dlion of Si~n~l Se~uence PCR amplification was used to generate an immunoglobulin signal sequence suitable for secretion of the B7-2Ig fusion protein from m~rnm~ n cells. The Ig signal sequence was WO 95t03408 2 ~ PCT/US94/08423 d from a plasmid cont~ining the murine IgG heavy chain gene (Orlandi. R. et al.
(1989) Proc. Natl. Acad. Sci. USA. 86:38333837) using the oligonucleotide 5'-GGCACTAGGTCTCCAGCTTGAGATCACAGTTCTCTCTAC-3' (#01) (SEQ ID NO: ) as the forward primer and the oligonucleotide 5'-GCTTGAATCTTCAGAGGAGCGGAGTGGACACCTGTGG-3' (#02) (SEQ ID NO: ) as the reverse PCR primer. The forward PCR primer (SEQ ID NO: ) contains recognition sequences for restriction enzymes BsaI and is homologous to sequences 5' to the initiating methionine of the Ig signal sequence. The reverse PCR primer (SEQ ID NO: ) is composed of sequences derived from the 5' end of the extracellular domain of hB7-2 and the 3' end of the Ig signal sequence. PCR amplification of the murine Ig signal template DNA using these primers resulted in a 224 bp product which is composed of BsaI restriction sites followed by the sequence of the Ig signal region fused to the first 20 nucleotides of the coding sequence of the extracellular domain of hB7-2. The junction between the signal sequence and hB7-2 is such that protein translation beginning at the signal sequence will continue into and through hB7-2 in the correct reading frame.

(2). Pl~a~dlion of thf? hP~7-2 Gene St~ ment The extracellular domain of the hB7.2 gene was prepared by PCR amplification of plasmid cont~inin~ the hB7-2 cDNA inserted into t;~lession vector pCDNAI (Freeman et al., Science 262:909-11 (1994)):
The extracellular domain of hB7-2 was prepared by PCR amplification using oligonucleotide 5'-GCTCCTCTGAAGATTCAAGC-3' (#03) (SEQ ID NO: ) as the fonvard primer and oligonucleotide 5'-GGCACTATGATCAGGGGGAGGCTGAGGTCC-3' (#04) (SEQ ID NO: ) as the reverse prirner. The forward PCR primer contained sequencescorresponding to the first 20 nucleotides of the B7-2 extracellular domain and the reverse PCR primer contained sequences corresponding to the last 22 nucleotides of the B7-2 extracellular domain followed by a Bcl I restriction site and 7 noncoding nucleotides. PCR
amplification with primer #03 and #04 yields a 673 bp product corresponding to the extracellular IgV and IgC like domains of hB7-2 followed by a unique Bcl I restriction site.
The signal sequence was attached to the extracellular portion of hB7-2 by PCR asfollows. DNA-PCR products obtained above corresponding to the signal sequence and the hB7-2 extracellular domain were mixed in equimolar amounts, denatured by heating to 100C, held at 54C for 30C to allow the complementary ends to anneal and the strands were filled in using dNTPs and Toq polymerase. PCR primers #01 and #04 were added and the entire fragment produced by PCR amplification to yield a ~880 fragment composed of a BsaI restriction site followed by the signal sequence fused to the extracellular domain of hB7-2, followed by a Bcl I restriction site.

9 ~

(3) Clonin~ ~nd Mo(lification of ~mmlmo~loblllin Fusion Do~in Plasmid pSP72 lgGI was prepared by cloning the 2000 bp segment of human IgGI
heavy chain genomic DNA (Ellison, J.W., et al. (1982) Nucl. Acids. Res. 10:4071-4079) into 5 the multiple cloning site of cloning vector pSP72 (Promega, Madison, Wl). Plasmid pSP721 gGI contained genomic DNA encoding the CHI, hinge, CH2 and CH3 domains of the heavy chain human IgC~1 gene. PCR primers designed to amplify the hinge-CH2-CH3 portion of the heavy chain along with the intervening DNA were pl~ed as follows. The forward PCR primer 5'-GCATTTTAAG( l~l-l l l CCTGATCAGGAGCCCAAATCTTCT
10 GACAAAACTCACACATCTCCACCGTCTCCAGGTAAGCC-3' (SEQ ID NO: ) contained HindIII and Bcl I restriction sites and was homologous to the hinge domain sequence except for five nucleotide substitutions which would change the three cysteine residues to serines. The reverse PCR primer 5'TAATACGACTCACTATAGGG-3' (SEQ ID
NO: ) was identical to the commercially available T7 primer (Promega, Madison, Wl).
Amplification with these primers yielded a 1050 bp fiagment bounded on the 5' end by HindIII and BclI restriction sites and on the 3' end by BamH1, Smal, Kpnl, Sacl, EcoR1, Clal, EcoR5 and Bglll restriction sites. This fragment contained the IgC hinge domain in which the three cysteine codons had been replaced by serine codons followed by an intron, the CH2 domain, an intron, the CH3 domain and additional 3' sequences. After PCR20 amplification, the DNA fragment was digested with Hindlll and EcoRl and cloned into c ~,es~ion vector pNRDSH digested with the same restriction enzymes. This created plasmid pNRDSH/IgG 1.
A similar PCR based strategy was used to clone the hinge-CH2-CH3 domains of hurnan IgCgamrna4 constant regions. A plasmid, p428D (Medical Research Council, 25 London, F.npl~n~) cont~ining the complete IgCgamma4 heavy chain genomic sequence (Ellison, J. Buxb~llnn, J. and Hood, L.E. (1981) DNA 1: 11 -18) was used as atemplate for PCR amplification using oligonucleotide 5'GAGCATTTTCCTGATCAGGA
GTCCAAATATGGTCCCCCATCCCATCATCCCCAGGTAAGCCAACCC-3' (SEQ ID
NO: ) as the forward PCR primer and oligonucleotide 30 S'GCAGAGGAATCGAGCTCGGTACCCGGGGATCCCCAGTGTGGGGACAGTGGGA
CCGCTCTGCCTCCC-3' (SEQ ID NO: ) as the reverse PCR primer. The forward PCR
- primer (SEQ ID NO: ) contains a Bcl l restriction site followed by the coding sequence for the hinge domain of IgCgamma4. Nucleotide substitutions have been made in the hinge region to replace the cysteines residues with serines. The reverse PCR primer (SEQ ID NO. ) 35 contains a PspAI restriction site (5'CCCGGG-3'). PCR amplification with these primers results in a 1179 bp DNA fragment. The PCR product was digested with Bcll and PspAI and ligated to pNRDSH/IgGl digested with the same restriction enzymes to yield plasmid wo g~/03408 ,~ 9 ~ PCTIUS94/08423 ~

;, -86-pNRDSH/IgG4. In this reaction, the IgCr 4 domain replaced the IgCyl domain present in pNRDSH/IgGl .
Modification of the CH2 domain in IgC to replace amino acids thought to be involved in binding to Fc receptor was accomplished as follows. Plasmid pNRDSH/IgGl served as template for modifications of the IgCrl CH2 domain and plasmid pNRDSH/IgG4 served as template for modifications of the IgC~ 4 CH2 domain. Plasmid pNRDSH/IgGl was PCRamplified using a fol~d PCR primer (SEQ ID NO: ) and oligonucleotide 5'-GGGTTTT
GGGGGGAAGAGGAAGACTGACGGTGCCCCC TCGGCTTCAGGTGCTGAGGAAG-3' (SEQ ID NO: ) as the reverse PCR primer. The forward PCR primer (SEQ ID NO: ) has been previously described and the reverse PCR primer (SEQ ID NO: ) was homologous to the amino termin~l portion of the CH2 domain of IgGl except for five nucleotide substitutions ~le~ignecl to change amino acids 234, 235, and 237 (Canfield, S. M. and Morrison, S. L. (1991) J. ~cp. Med. 173: 1483-1491.) from Leu to Ala, Leu to Glu, and Gly to Ala, respectively. Amplification with these PCR primers will yield a 239 bp DNA
fragment con~i~ting of a modified hinge domain, an intron and modified portion of the CH2 domain. Plasmid pNRDSHlIgGl was also PCR amplified with the oligonucleotide 5'-CATCTCTTCCTCAGCACCTGAAGCCGAGGGGGCACCGTCAGTCTTCCTCTTCCC
CC-3' (SEQ ID NO: ) as the forward primer and oligonucleotide (SEQ ID NO: ) as the reverse PCR primer. The forward PCR primer (SEQ ID NO: ) is complemf?nt~ry to primer (SEQ ID NO: ) and contains the five complement~ry nucleotide changes necessary for the CH2 amino acid repl~ement~ The reverse PCR primer (SEQ ID NO: ) has been previously described. Amplification with these primes yields a 875 bp fragment con~i~ting of the modified portion of the CH2 domain, an intron, the CH3 domain, and 3' additional sequences.
The complete IgC~l segment consisting of modified hinge domain, modified CH2 domain and CH3 domain was prepared by an additional PCR reaction. The purified products of the two PCR reactions above were mixed, denatured (95C,1 minute) and then renatured (54C, 30 seconds) to allow complementary ends of the two fr~gment~ to anneal. The strands were filled in using dNTP and Taq polymerase and the entire fragment arnplified using forward PCR primer (SEQ ID NO: ) and reverse PCR primer (SEQ ID NO: ). The resulting fragment of 1050 bp was purified, digested with HindIII and EcoR1 and ligated to pNRDSH
previously digested with the same restriction enzymes to yield plasmid pNRDSHIgGl m.
Two amino acids at immllnoglobulin positions 235 and 237 were changed from Leu to Glu and Gly to Ala, respectively, within the IgCr4 CH2 domain to elimin~te Fc receptor binding. Plasmid pNRDSH/IgG4 was PCR amplified using the forward primer (SEQ ID NO:
) and the oligonucleotide 5'-CGCACGTGACCTCAGGGGTCCGGGAGATCATGAGAGTGTCCTTGGGTTTTGGGG
GGAACAGGAAGACTGATGGTGCCCCCTCGAACTCAGGTGCTGAGG-3 ' (SEQ ID

~0 95/03408 ~ l 6 ~ O 9 1 PCT/US94tO8423 NO: ) as the reverse primer. The forward primer has been previously described and the reverse primer was homologous to the amino terminal portion of the CH2 domain, except for three nucleotide substitutions designed to replace the amino acids described above. This primer also contained a Pmll restriction site for subsequent cloning. Amplification with these 5 primers yields a 265 bp fragment composed of the modified hinge region, and intron, and the modified 5' portion of the CH2 domain.
Plasmid pNRDSH/lgG4 was also PCR amplified with the oligonucleotide S
'-CCTCAGCACCTGAGTTCGAGGGGGCACCATCAGTCTCCTGTTCCCCCC
AAAACCCAAGGACACTCTCATGATCTCCCGGACCCCTGAGGTCACGTGCG-3 ' 10 (SEQ ID NO: ) as the forward primer and oligonucleotide (SEQ ID NO: ) as the reverse PCR
primer. The forward PCR primer (SEQ ID NO: ) is complement~ry to primer (SEQ ID NO: ) and contains the three complementary nucleotide changes necessary for the CH2 amino acid replacements. The reverse PCR primer (SEQ ID NO: ) has been previously described.
Amplification with these primes yields a 1012 bp fragment consisting of the modified portion 15 of the CH2 domain, an intron, the CH3 ~lom~in, and 3' additional sequences. The complete IgC~4 segment consisting of modified hinge domain, modified CH2 domain and CH3 domain was prepared by an additional PCR reaction. The purified products of the two PCR reactions above were mixed, denatured (95C,1 minute) and then renatured (54C, 30 seconds) to allow complementary ends of the two fragments to anneal. The strands were filled in using dNTP
20 and Taq polymerase and the entire fragment amplified using forward PCR primer (SEQ ID
NO: ) and reverse PCR primer (SEQ ID NO: ). The resulting fragment of 1179 bp was purified, digested with Bcll and PspAI and ligated to pNRDSH previously digested with the same restriction enzymes to yield plasmid pNRDSH/IgG4m.

(4). A~sernhly of Fin~l hP~7~ Genes The PCR fragment corresponding to the Ig signal-hB7-2 gene fusion prepared abovewas digested with BsaI and Bcl 1 restriction enzymes and ligated to pNRDSH/IgGl,pNRDSH/lgGlm, pNRDSH/IgG4, and pNRDSH/IgG4m previously digested with Hind III
and BclI. The ligated plasmids were transformed into E. coli JMI09 using CaC12 competent cells and transformants were selected on L-agar cont~ining ampicillin (50 ,ug/ml; Molecular Cloning: A Laboratory Manual (1982) Eds. M~ni~ti~, T., Fritsch, E. E., and Sambrook, J.
Cold Spring Harbor Laboratory). Plasmids isolated from the transformed E coli were analyzed by restriction enzyme digestion. Plasmids with the expected restriction plasmid were sequenced to verify all portions of the signal-hB7-2-IgG gene fusion segments.

wo 95/03408 ~ 7 ~ ~ ~ PCTIUS94/08423 n. Fxpression Clonin~ of h~7-2V-I~G 1 and hB7-2C I~G1 The variable and constant domains of human B7-2 were separately cloned into pNRDSH/IgG1. These clonings were accomplished using PCR. The portions of hB7-2 corresponding to the variable and constant regions were determined from intron/exon 5 mapping and previously published gene structure analysis.

Human B7-2 Variable Domain 5'GCTCCTCTGAAGATT......... GAACTGTCAGTGCTT3' (SEQ ID NO: ) A P L K I E L S V L (SEQ ID NO: ) Human B7-2 Constant Domain 5'GCTAACTTCAGTCAA......... CCTTTCTCTATAGAG3' (SEQ ID NO: ) A N F S Q P F S I E (SEQ ID NO: ) (1). ~sçnlbly of hB7-2VI~
The hB7-2V domain Ig sequence was assembled using a PCR strategy similar to thatshown above. The signal sequence was derived from the onco M gene by PCR amplification of a plasmid cont~inin~ the onco M gene using oligonucleotide 5'-GCAACCGGAAGCTTGCCACCATGGGGGTACTGCTCACACAGAGGACG-3' (#05) 20 (SEQ ID NO: ) as the forward PCR primer and 5'-AGTCTCATTGAAATAAGCTTGAATCTTCAGAGGAGCCATGCTGGCCATGCTTGGA
AACAGGAG-3' (#06) (SWQ ID NO: ) as the reverse primer. The forward PCR primer (#05) contains a Hind III restriction site and the amino t~rmin~l portion of the onco M signal sequence. The reverse PCR (#06) contains the sequence corresponding to the 3' portion of 25 the onco M signal sequence fused to the 5' end of the hB7-2 IgV like domain.
The hB7-2 IgV like domain was obtained by PCR amplification of the hB7-2 cDNA
using oligonucleotide 5'-CTCCTGTTTCCAAGCATGGCCAGCATGGCTCCTCTGAA
GATTCAGGCTTATTTCAATGAGAC-3' (#07) (SEQ ID NO: ) as the forward and oligonucleotide 5'-30 TGTGTGTGGAATTCTCATTACTGATCAAGCACTGACAGTTCAGAATTCATC-3' (#08) (SEQ ID NO: ) as the reverse PCR primer. PCR amplification with these primers yields the hB7-2 IgV domain with a portion of the 3' end of the onco M signal sequence on the 5' end and a Bcl I restriction site on the 3' end. The signal and IgV domain were linked together in a PCR reaction in which equimolar amounts of the onco M signal and IgV domain 35 DNA fragments were mixed, denatured, annealed, and the strands filled in. Subsequent PCR
amplification using forward primer #05 and reverse primer #08 yielded a DNA fragment co~ i"i"g a Hind III restriction site, followed by the onco M signal fused to the B7-2 IgV

~O 95l03408 1 ~ 7 0 9 1 PCT/US94/08423 domain followed by a Bcl I restriction site. This PCR fragment was digested with Hind II
and Bcl I and cloned into expression vector pNRDSH/IgG1 digested with the same restriction enzymes to yield pNRDSH/B7-2CIg.

(2). A~mbly of hR7-~CI~
The ~ s~ion plasmid for hB7-2IgC domain was prepared as described above for the IgV domain except for using PCR primers specific for the IgC domain. The onco M
signal sequence was prepared using oligonucleotide #05 as the forward PCR primer and oligonucleotide 5'-AACAGGAG-3' (#09) (SEQ ID NO: ) as the reverse PCR primer. The hB7-2 IgC domain was prepared using oligonucleotide 5'-CTCCTGTTTCCAAGCATGGCCAGCATGGCTAACTTCAGTC
AACCTGAAATAGTACCAATTTC-3' (#11) (SEQ ID NO: ) as the reverse PCR primer.
The two PCR products were mixed and amplified with primers #05 and #1 1 to assemble the onco M signal sequence with the hB7-2IgC domain. The PCR product was subsequently digested with Hind III and BclI and ligated to pNRDSH/IgG1 digested with similarrestriction enzymes to yield the final ex~r~ssion plasmid pNRDSH~B7-2CIgG1.

F Cornr-etition Rintli~ Ac~ys With Hl~m~n ~7-~Tg Fusion Prote;n~
The ability of various B7 farnily-Ig fusion proteins to competitively inhibit the binding of biotinylated-CTLA4Ig to immobilized B7-2Ig was determinç-1 Competition binding assays were done as follows and analysed according to McPherson (McPherson, G.A. (1985) J. Pharmacol. Methods 14:213-228). Soluble hCTLA4Ig was labelled with 125I
to a specific activity of approximately 2 x 1 o6 cpm/pmol. hB7-2-Ig fusion protein was coated overnight onto microtiter plates at l0~Lg/ml in 10 mM Tris-HCl, pH8.0, 50 ,ul /well.
The wells were blocked with binding buffer (DMEM cont~ining 10% heat-inactivated FBS, 0.1% BSA, and 50 mM BES, pH 6.8) for 2 h at room temperature. The labeled CTLA4-Ig (4nM) was added to each well in the presence or absence of unlabeled competing Ig fusion proteins, including full-length B7-2 (hB7-2Ig), full-length B7-1 (hB7-lIg), the variable region-like domain of B7-2 (hB7-2VIg) and the constant region-like domain of B7-2 (hB7-- 2~Ig) and allowed to bind for 2.5 h at room temperature. The wells were washed once with ice-cold binding buffer and then four times with ice-cold PBS. Bound radioactivity was recovered by treatment of the wells with 0.5 N NaOH for 5 min and the solubilized material removed and counted in a gamma counter.
The results of these assays are shown in Figure 15 in which both hB7-2Ig (10-20 nM) and hB7-2VIg (30-40 nM) competitively inhibit the binding of CTLA4Ig to immobilized B7-WO 95/03408 PCTIUS94/08423 ~
~7~
so-2 protein. hB7-2CIg is unable to compete with soluble CTLA4, indicating that the B7-2 binding region is in found in the variable-region like domain.

F. Competitive bindin~ Assays for B7-1 and B7-2 fusion proteins The ability of the various recombinant CTLA4 forms to bind to hB7-1 or hB7-2 was~s~essed in a competitive binding ELISA assay as follows. Purified recombinant hB7-Ig (20 ~Lg/ml in PBS) was bound to a Costar EIAIRIA 96 well microtiter dish (Costar Corp, Cambridge MA, USA) in 50 ~L overnight at room temperature. The wells were washed three times with 200 ~lL of PBS and the unbound sites blocked by the addition of 1 % BSA in PBS
(200/well) for 1 hour at room temperature. The wells were washed as above. Biotinylated hCTLA4IgG1 (ref, MFGR;1 ~Lg/ml serially diluted in twofold steps to 15.6 ng/mL; 50 ,uL) was added to each well and incubated for 2.5 hours at room tc;lllp~ld~-lre. The wells were washed as above. The bound biotinylated CTLA4Ig was detected by the addition of 50 1/l of a 1 :2000 dilution of streptavidin-HRP (Pierce Chemical Co., Rockford, IL) for 30 minutes at room temperature. The wells were washed as above and 50 ,uL of ABTS (Zymed, California) added and the developing blue color monitored at 405 nm after 30 min. A graphic representation of a typical binding assay is shown in Figure 16. The ability of the various forms of CTLA4 to compete with biotinylated CTLA4IgG1 was ~se~ecl by mixing varying amounts of the competing protein with a quantity of biotinylated CTLA4IgGl shown to be non-saturating (i.e., 70 ng/mL; 1.5nM) and perfor~ning the binding assays as described above (Figure 15). A reduction in the signal (Abs 405 nm) expected for biotinylated CTLA4IgGl indicated a competition for binding to hB7-1.
Considering the previous evidence that CTLA4 was the high affinity receptor for B7-1, the avidity of binding of CTLA4 and CD28 to B7-1 and B7-2 was compared. B7-1-Ig or B7-2-Ig was labelled with biotin and bound to immobilized CTLA4-Ig in the presence or absence of increasing concentrations of unlabeled B7-1-Ig or B7-2-Ig. The experiment was repeated with 125-I-labeled B7-1-Ig or B7-2-Ig. Using this solid phase binding assay, the avidity of B7-2 (2.7 nM) for CTLA4 was dete~nined to be approximately two-fold greater ~an that observed for B7-1 (4.6 nM). The experimentally determined ICso values are indicated in the upper right corner of the panels. The affinity of both B7-1 and B7-2 for CD28 was lower and was difficult to confidently determine.

?~ ~7~gl ~0 95/03408 ~ PCT/US94/08423 Production ~nd Characteri~ion of Monoclonal Antibo~lies to Hllm~n n7-2 ~. Tmmllni7~tion~ ~nd Cell Fusio}l~
Balb/c female mice (obtained from Taconic Labs, Germantown, NY) were immllni7~d hlll~p~liLoneally with 50 ,~Lg human B7.2-Ig emulsified in complete Freund's adjuvant (Sigma Chemical Co., St. Louis, MO) or 106 CHO-human B7.2 cells per mouse. The mice were given two booster immlmi7~tions with 10-25 ~lg human B7.2-Ig emulsified in incomplete Freund's adjuvant (Sigma Chemical Co., St. Louis, MO) or CHO-human B7.2 cells at10 fourteen, day intervals following the initial i~mmunization for the next two months. The mice were bled by retro-orbital bleed and the sera assayed for the presence of antibodies reactive to the immllnogen by ELISA against human B7.2-Ig. ELISA against hCTLA4-Ig was also used to control for Ig tail directed antibody responses. Mice showing a strong serological response were boosted intravenously via the tail vein with 25 ~g human hB7.2-Ig diluted in phosphate-15 buffered saline (PBS), pH 7.2 (GIBCO, Grand Island, NY). Three to four days following this boost, the spleens from these mice were fused 5:1 with SP 2/0 myeloma cells (American Type Culture Collection, Rockville, MD, No. CRL8006), which are inc~p~kle of secreting both heavy and light immunoglobulin chains (Kearney et al. (1979) J. Immunol. l ~3: 1548).
Standard methods based upon those developed by Kohler and Milstein (Nature (1975) 20 ~:495) were used.

F3. Antibo~y Screenir~
After 10-21 days, supern~t~nt~ from wells cont~ining hybridoma colonies from thefusion were screened for the presence of antibodies reactive to human B7.2 as follows: Each 25 well of a 96 well flat bottomed plate (Costar Corp., Cat.3~3590) was coated with 50 ~11 per well of a I ,ug/ml hurnan B7.2-Ig solution or S x 104 3T3-hB7.2 cells on Iysine coated plates in phosphate-buffered saline, pH 7.2, overnight at 4 C. The hurnan B7.2-Ig solution was aspirated off, or the cells were cross-linked to the plates with glutaraldehyde, and the wells were washed three times with PBS, then blocked with 1% BSA solution (in PBS) (1001l 30 l/well) for one hour at room temperature. Following this blocking incubation, the wells were washed three times with PBS and 50 1ll of hybridoma supernatant was added per well and incubated for 1.5 hours at room temperature. Following this incubation, the wells were washed three times with PBS and then incubated for 1.5 hours at room temperature with 50 ~Ll per well of a I :4000 dilution of horseradish peroxidase-conjugated, affinity purified, goat 35 anti-mouse IgG or IgM heavy and light chain-specific antibodies (HRP; Zymed Laboratories, San Francisco, CA). The wells were then washed three times with PBS, followed by a 30 minute incubation in 50 ,ul per well of I mM 2,2-azino-bis-3-ethylben7t~ 7Oline-6-sulfonic wO g~/03408 2~ 9~ PCT/US94/08423 acid (ABTS) in 0.1 M Na-Citrate, pH 4.2 to which a 1:1000 dilution of 30 % hydrogen peroxide had been added as a substrate for HRP to detect bound antibody. The absorbence was then deterrnined at OD410 on a spectrophotometric autoreader (Dynatech, Virginia).
Three hybridomas, HA3.1F9, HA5.2B7 and HF~.3Dl, were identified that produced antibodies to human B7.2-Ig. HA3.1F9 was determined to be of the IgGl isotype, HA5.2B7 was determined to be of the IgG2b isotype and HF2.3Dl as deterrnined to be of the IgG2a isotype. Each of these hybridomas were subcloned two additional times to insure that they were monoclonal. Hybidoma cells were deposited with the American Type Culture Collection, which meets the requirements of the Budapest Treaty, on July 19, 1994 as ATCC
Accession No. (hybridoma HA3.1F9), ATCC Accession No. (HA5.2B7) and ATCC Accession No. (HF2.3D1).

C. Colnpetitive FT ISA
Supçrn~t~nt~ from the hybridomas HA3.1F9, HA5.2B7 and HF2.3Dl were further characterized by competitive ELISA, in which the ability of the monoclonal antibodies to inhibit the binding of biotinylated hCTLA4Ig to immobilized hB7-2 immlln~globulin fusion proteins was e~r~mined. Biotinylation of hCTLA4Ig was perforrned using Pierce ~mml-nopure NHS-LC Biotin (Cat. No. 21335). B7-2 immunoglobulin fusion proteins used were: hB7.2-Ig (full-length hB7-2), hB7.2-VIg (hB7-2 variable domain only) and hB7.2-CIg (B7-2 constant domain only). ~ hB7.1 -Ig fusion protein was used as a control. For the ELISA, 96 well plates were coated with the Ig fusion protein (50 ,~Ll/well of a 20 ~Lg/ml solution) overnight at room tc~ c~ . The wells were washed three times with PBS,blocked with 10 % fetal bovine serum (FBS), 0.1 % bovine serum albumin (BSA) in PBS for 1 hour at room temperature, and washed again three times with PBS. To each well was added 50 ~1 of Bio-hCTLA4-Ig (70 ng/ml) and 50 ,ul of competitor monoclonal antibody supernatant. Control antibodies were an anti-B7.1 mAb (EW3.5D12) and the anti-hB7-2 mAb B70 (IgG2bK, obtained from Ph~rmin~en). The wells were washed again and streptavidin-conjugated horse radish peroxidase (from Pierce, Cat. No. 21126; 1 :2000 dilution, 50 ,ul/well) was added and incubated for 30 minutes at room tem~c,alllre. The wells were washed again, followed by a 30 minute incubation in 50 ~11 per well of ABTS in 0.1 M
Na-Citrate, pH 4.2 to which a 1:1000 dilution of 30 % hydrogen peroxide had been added as a substrate for HRP to detect bound antibody. The absorbence was then determined at OD410 on a spectrophotometric autoreader (Dynatech, Virginia). The results, sho~-vn in Table IV below, demonstrate that each of the mAbs produced by the hybridomas HA3.1 F9, HA5.2B7 and HF2.3D1 are able to co,,l~clilively inhibit the binding of hCLTA4Ig to full-length hB7.2-Ig or hB7.2-VIg (hCTLA4Ig does not bind to hB7.2CIg).

2 ~ 9 1 ~tO 95/03408 l'CT/US9~/08423 TART ~ IV

Blocking of F3in~
hR7.1 -1~ hR7.2-I~ hR7.2-VIg hR7.2-CIg EW3.5Dl2 (anti-hB7.1 mAb) Yes No No No B70 (anti-hB7-2) No Yes Yes No HA3.1F9 (anti-hB7-2) No Yes Yes No HA5.2B7 (anti-hB7-2) No Yes Yes No HF2.3D1 (anti-hB7-2) No Yes Yes No 5 r). Flow Cytometry Supernatants from the hybridomas HA3.1F9, HA5.2B7 and HF2.3D1 were also characterized by flow cytometry. Supern~t~ntc collected from the clones were screened by flow cytometry on CHO and 3T3 cells transfected to express hB7.2 (CHO-hB7.2 and 3T3-h~7.2, respectively) or control transfected 3T3 cells (3T3-Neo). Flow cytometry was performed as follows: 1 x 106 cells were washed three times in 1 % BSA in PBS, then the cells were incubated in 50,ul hybridoma supernatant or culture media per 1 x I o6 cells for 30 minutes at 4 C. Following the incubation, the cells were washed three times with l % BSA
in PBS, then incubated in 50 ,ul fluorescein-conjugated goat anti-mouse IgG or IgM
antibodies (Zymed Laboratories, San Francisco, CA) at 1 :50 dilution per 1 x 1 o6 cells for 30 15 rninlltes at 4 C. The cells were then washed three times in 1 % BSA in PBS and fixed with 1 % p~aro..naldehyde solution. The cell sarnples were then analyzed on a FACScan flow cytometer (Becton Dickinson, San Jose CA). The results, shown in Figures 17, 18 and 19, demonstrate the monoclonal antibodies produced by the hybridomas HA3.1F9, HA5.2B7 and HF2.3Dl each bind to hB7-2 on the surface of cells.
F Inhibition of Prolifer~tio~ of H--m~n T Cell~ by Anti-hR7-2 rn~bs Hybridoma supern~t~nt~ cont~ining anti-hurnan B7-2 mAbs were tested for their ability to inhibit hB7-2 costimulation of human T cells. In this assay, purified CD28+ human T cells were treated with submitogenic amounts of PMA (lng/ml) to deliver the primary signal and with CHO cells expressing hB7-2 on their surface to deliver the costimulatory signal. Proliferation of the T cells was measured after three days in culture by the addition of 3H-thymidine for the rem~inin~ 18 hours. As shown in Table V, resting T cells show little proliferation as measured by 3H-thymidine incorporation (510 pm). Delivery of signal 1 by PMA results in some proliferation (3800 pm) and T cells receiving both the primary (PMA) and costimulatory (CHO/hB7-2) signals proliferate m~im~lly (9020 cpm). All three anti-WO95/03408 ~ PCT/US94/08423 hB7-2 mAbs tested reduce the costimulatory signal intlllcecl proliferation to that found for PMA treated cells alone showing that these mAbs can inhibit T cell proliferation by blocking the B7/CD28 costimulatory pathway.

TARTFV

Addition to CD28+ T Cells hB7-2 mAb CPM

+PMA --- 3800 +PMA + CHOlhB7-2 --- 9020 +PMA + CHO/hB7-2 HF2.301 3030 --- HA5.2B7 1460 --- HA3.1F9 2980 10Re~ressior~ of I~planted T--mor Cells Transfected to F.~?ress ~7-2 In this example, untransfected or B7-2 transfected J558 plasmacytoma cells were used in turnor regression studies to exAmine the effect of ~ cssion of B7-2 on the surface of tumor cells on the growth of the tumor cells when transplanted into ~nimAI~
15J558 plasmacytoma cells (obtained from the American Type Culture Collection, Rockville, MD; # TIB 6) were transfected with an expression vector cont~ining cDNA
encoding either mouse B7-2 (pAWNE03) or B7-1 (pNRDSH or pAWNE03) and a neomycin-resi~tAnce gene. Stable trAn~fectAnt~ were selected based upon their neomycin resictAnce and cell surface expression of B7-2 or B7-1 on the tumor cells was confirme~l by FACS analysis 20 using either an anti-B7-2 or anti-B7-1 antibody.
Syngeneic Balb/c mice, in groups of 5-10 mice/set, were used in experiments ~lesi~n~d to determine whether cell-surface expression of B7-2 on tumor cells would result in regression of the implanted tumor cells. Untransfected and transfected J558 cells were cultured in vitro, collected, washed and resuspended in Hank's buffered salt solution 25 (GIBCO, Grand Island, New York) at a concentration of 10~ cells/ml. A patch of skin on the right flank of each mouse was removed of hair with a depilatory and, 24 hours later, 5 x 106 tumor cells/mouse were implanted intradermally or subdermally. Measurements of tumor volume (by linear measurements in three perpendicular directions) were made every two to three days using calipers and a ruler. A typical experiment lasted 18-21 days, after which ~O 95/0340~ ~ 1 6 7 0 91 PCTIUS94/08423 time the tumor size exceeded 10 % of the body mass of mice transplanted with untransfected, control J558 cells. As shown in Figure 20, J558 cells transfected to express B7-2 on their surface were rejected by the mice. No tumor growth was observed even after three weeks.
Similar results were observed with J558 cells transfected to express B7-1 on their surface. In 5 contrast, the untransfected (wild-type) J558 cells produced massive tumors in as little as 12 days, requiring the animal to be enth~ni7~cl This example demonstrates that cell-surface expression of B7-2 on tumor cells, such as by transfection of the tumor cells with a B7-2 cDNA, induces an anti-tumor response in naive ~nim~l~ that is sufficient to cause rejection of the tumor cells.
FQUIVAT .FNTS
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.

WO95/03408 2~6~9~ PCTIUS94/08423 ~

SEQUENCE LISTING
(1) GENERAL INFORMATION:
(i) APPLICANT:
(A) NAME: DANA-FARBER CANCER lNS'l'l'l'U'l'~
(B) STREET: 44 BINNEY STREET
(C) CITY: BOSTON
(D) STATE: MASSA~u~LlS
~ (E) COUNTRY: USA
= (F) POSTAL CODE (ZIP): 02115 (G) TELEPHONE: (617) 632-4016 (H) TELEFAX: (617) 632-4012 (A) NAME: REPLIGEN CORPORATION
(B) STREET: ONE KENDALL SQUARE, BLDG 700 (C) CITY: CAMBRIDGE
(D) STATE: MASSA~US~llS
(E) COUNTRY: USA
(F) POSTAL CODE (ZIP): 02139 (G) TELEPHONE: (617) 225-6000 (H) TELEFAX: (617) 494-1975 (ii) TITLE OF INVENTION: Novel CTLA4/CD28 Ligands and Uses Therefor (iii) NUMBER OF SEQUENCES: 31 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: LA~IVE ~ COCKFIELD
(B) STREET: 60 State Street, Suite 510 (C) CITY: Boston (D) STATE: Massachusetts (E) COUNL~Y: USA
(F) ZIP: 02109 (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (c) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release #1.0, Version ~1.25 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER:
4~ (B) FILING DATE:
(C) CLASSIFICATION:
(vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US08/101,624; US08/109,393; US08/147,773 (B) FILING DATE: 26-JUL-1993; 19-AUG-1993; 03-NOV-1993 (viii) ArlloKN~y/AGENT INFORMATION:
(A) NAME: Mandragourasl Amy E.
(B) REGISTRATION NUMBER: 36,207 (C) REFERENCE/DOCKET NUMBER: RPI-004CP2PC

~vo 95~03408 2 ~ 1 PCT/US94/08423 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (617) 227-7400 (B) TELEFAX: (617) 227-5941 - (2) INFORMATION FOR SEQ ID NO:1:
~ Q~ ~ CHARACTERISTICS:
(A) LENGTH: 1120 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear ( ii ) MOLECULE TYPE: cDNA
(ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 107..1093 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

Met Asp Pro Gln Cys Thr Met Gly Leu Ser Asn Ile Leu Phe Val Met Ala Phe Leu 35 Leu Ser Gly Ala Ala Pro heu Lys Ile Gln Ala Tyr Phe Asn Glu Thr Ala Asp Leu Pro Cys Gln Phe Ala Asn Ser Gln Asn Gln Ser Leu Ser Glu Leu Val Val Phe Trp Gln Asp Gln Glu Asn Leu Val Leu Asn Glu Val Tyr Leu Gly Lys Glu Lys Phe Asp Ser Val His Ser Lys Tyr Met Gly Arg Thr Ser Phe Asp Ser Asp Ser Trp Thr Leu Arg Leu His Asn WO 95/03408 2 ~ 1 PCT/US94/08423 CTT CAG ATC AAG GAC AAG GGC TTG TAT CAA TGT ATC ATC CAT CAC A~A 451 Leu Gln Ile Lys Asp Lys Gly Leu Tyr Gln Cys Ile Ile His His Lys AAG CCC ACA GGA ATG ATT CGC ATC CAC CAG ATG A~T TCT GAA CTG TCA 499 Lys Pro Thr Gly Met Ile Arg Ile His Gln Met Asn Ser Glu Leu Ser 0 Val Leu Ala Asn Phe Ser Gln Pro Glu Ile Val Pro Ile Ser Asn Ile Thr Glu Asn Val Tyr Ile Asn Leu Thr Cys Ser Ser Ile His Gly Tyr Pro Glu Pro Lys Lys Met Ser Val Leu Leu Arg Thr Lys Asn Ser Thr ATC GAG TAT GAT GGT ATT ATG QG A~A TCT CAA GAT AAT GTC ACA GAA 691 Ile Glu Tyr Asp Gly Ile Met Gln Lys Ser Gln Asp Asn Val Thr Glu Leu Tyr Asp Val Ser Ile Ser Leu Ser Val Ser Phe Pro Asp Val Thr Ser Asn Met Thr Ile Phe Cy5 Ile Leu Glu Thr Asp Lys Thr Arg Leu Leu Ser Ser Pro Phe Ser Ile Glu Leu Glu Asp Pro Gln Pro Pro Pro Asp His Ile Pro Trp Ile Thr Ala Val Leu Pro Thr Val Ile Ile Cys GTG ATG GTT TTC TGT CTA ATT CTA TGG A;~A TGG AAG AAG AAG AAG CGG 931 Val Met Val Phe Cys Leu Ile Leu Trp Lys Trp Lys Lys Lys Lys Arg CCT CGC AAC TCT TAT A~A TGT GGA ACC AAC ACA ATG GAG AGG GAA GAG 979 Pro Arg Asn Ser Tyr Lys Cys Gly Thr Asn Thr Met Glu Arg Glu Glu Ser Glu Gln Thr Lys Lys Arg Glu Lys Ile His Ile Pro Glu Arg Ser GAT GAA GCC CAG CGT GTT TTT A~A AGT TCG AAG ACA TCT TCA TGC GAC 1075 Asp Glu Ala Gln Arg Val Phe Lys Ser Ser Lys Thr Ser Ser Cys Asp *VO 95/03408 21 ~ ~ O 91 PCTIUS94/08423 _99 Lys Ser Asp Thr Cys Phe s (2) INFORMATION FOR SEQ ID NO:2:
(i) SEQUENCE CHARACTBRISTICS:
(A) LENGTH: 329 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:
Met Asp Pro Gln Cys Thr Met Gly Leu Ser Asn Ile Leu Phe Val Met Ala Phe Leu Leu Ser Gly Ala Ala Pro Leu Lys Ile Gln Ala Tyr Phe Asn Glu Thr Ala Asp Leu Pro Cys Gln Phe Ala Asn Ser Gln Asn Gln Ser Leu Ser Glu Leu Val Val Phe Trp Gln Asp Gln Glu Asn Leu Val Leu Asn Glu Val Tyr Leu Gly Lys Glu Lys Phe Asp Ser Val His Ser Lys Tyr Met Gly Arg Thr Ser Phe Asp Ser Asp Ser Trp Thr Leu Arg Leu His Asn Leu Gln Ile Lys Asp Lys Gly Leu Tyr Gln Cys Ile Ile His His Lys Lys Pro Thr Gly Met Ile Arg Ile His Gln Met Asn Ser Glu Leu Ser Val Leu Ala Asn Phe Ser Gln Pro Glu Ile Val Pro Ile Ser Asn Ile Thr Glu Asn Val Tyr Ile Asn Leu Thr Cys Ser Ser Ile His Gly Tyr Pro Glu Pro Lys Lys Met Ser Val Leu Leu Arg Thr Lys Asn Ser Thr Ile Glu Tyr Asp Gly Ile Met Gln Lys Ser Gln Asp Asn Val Thr Glu Leu Tyr Asp Val Ser Ile Ser Leu Ser Val Ser Phe Pro WO 95/03408 2~ PCT/US94/08423 ~

Asp Val Thr Ser Asn Met Thr Ile Phe Cys Ile Leu Glu Thr Asp Lys 5 Thr Arg Leu Leu Ser Ser Pro Phe Ser Ile Glu Leu Glu Asp Pro Gln Pro Pro Pro Asp His Ile Pro Trp Ile Thr Ala Val Leu Pro Thr Val Ile Ile Cys Val Met Val Phe Cys Leu Ile Leu Trp Lys Trp Lys Lys Lys Lys Arg Pro Arg Asn Ser Tyr Lys Cys Gly Thr Asn Thr Met Glu Arg Glu Glu Ser Glu Gln Thr Lys Lys Arg Glu Lys Ile His Ile Pro Glu Arg Ser Asp Glu Ala Gln Arg Val Phe Lys Ser Ser Lys Thr Ser Ser Cys Asp Lys Ser Asp Thr Cys Phe (2) INFORMATION FOR SEQ ID NO:3:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 20 base pairs tB) TYPE: nucleic acid (C) sTRANn~nN~s single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

(2) INFORMATION FOR SEQ ID NO:4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) sTR~Nn~n~s single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

o 9 ~
, ~ 0 95/03408 PCTrUS94/08423 TAAGGTTCCT TCACA~AG 18 ~2) INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) sTRp~n~n~cs single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) ~.Qu~ DESCRIPTION: SEQ ID NO:5:

(2) INFORMATION FOR SEQ ID NO:6:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRPNn~nN~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:
ATGCGA~TCA TTCCTGTGGG C 21 (2) INFORMATION FOR SEQ ID NO:7:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRA~N~:SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:

(2) INFORMATION FOR SEQ ID NO:8:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid WO 95/03408 ~ . PCT/US94/08423 (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide S

(xi) ~u~N~ DESCRIPTION: SEQ ID NO:8:
10 CTCTCAAAAC CA~AGCCTGA G 21 (2) INFORMATION FOR SEQ ID NO:9:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid ~ (C) sTR~N~n~s single = (D) TOPOLOGY: linear - 20 (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:

(2) INFORMATION FOR SEQ ID NO:10:
( i ) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STR~NDEDNESS: 8 ingle (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) ~U~N~ DESCRIPTION: SEQ ID NO:10:
TCTGGA~ACT GACAAGACGC G 21 (2) INFORMATION FOR SEQ ID NO:11:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide ~VO 95/03408 ~ PCT/US94/08423 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:11:

5 (2) INFORMATION FOR SEQ ID NO:12:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12:
CA~l-lcllC CCTCTCCATT G 21 (2) INFORMATION FOR SEQ ID NO:13:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANu~uN~SS: 5 ingle (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13:

(2) INFORMATION FOR SEQ ID NO:14:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) ShQU~. DESCRIPTION: SEQ ID NO:14:

(2) INFORMATION FOR SEQ ID NO:15:
( i ) S~u~N~ CHARACTERISTICS:

WO 95/03408 21~ 7 0 ~ ~ PCT/US94/08423 ~

~A) LENGTH: 12 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: 5 ingle (D) TOPOLOGY: linear tii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:15:

(2) INFORMATION FOR SEQ ID NO:16:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 8 base pairs (B) TYPE: nucleic acid (C) STR~ S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(xi) ~u~N~ DESCRIPTION: SEQ ID NO:16:
CTCTA~AG 8 (2) INFORMATION FOR SEQ ID NO:17:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:17:
Lys Tyr Met Gly Arg Thr Ser Phe Asp (2) INFORMATION FOR SEQ ID NO:18:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 13 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide ~WO 95t03408 ~ 16 ~ ~ 91 PCTtUS94tO8423 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:18:
Lys Ser Gln Asp Asn Val Thr Glu Lys Tyr Asp Val Ser (2) INFORMATION FOR SEQ ID NO:19:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 15 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:19:
Trp Lys Trp Lys Lys Lys Lys Arg Pro Arg Asn Ser Tyr Lys Cys (2) INFORMATION FOR SEQ ID NO:20:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) sTRANn~nN~s single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (Xi) ~QU~N~'~ DESCRIPTION: SEQ ID NO:20:

(2) INFORMATION FOR SEQ ID NO:21:
( i ) ~QU~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (c) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleo~ide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:21:

WO 95/03408 PCTIUS94/08423 ~
2 ~

(2) INFORMATION FOR SEQ ID NO:22:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1163 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear 10 (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 111.. 1040 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:22:
~ 20 CCCACGCGTC CGGGAGCAAG CAGACGCGTA AGAGTGGCTC CTGTAGGCAG CACGGACTTG 60 - AACAACCAGA CTCCTGTAGA C~L~llC~AG AACTTACGGA AGCACCCACG ATG GAC 116 Met Asp Pro Arg Cys Thr Met Gly Leu Ala Ile Leu Ile Phe Val Thr Val Leu Leu Ile Ser Asp Ala Val Ser Val G1U Thr Gln Ala Tyr Phe Asn Gly 35 Thr Ala Tyr Leu Pro Cys Pro Phe Thr Lys Ala Gln Asn Ile Ser Leu Ser Glu Leu Val Val Phe Trp Gln Asp Gln Gln Lys Leu Val Leu Tyr GAG CAC TAT TTG GGC ACA GAG A~A CTT GAT AGT GTG AAT GCC AAG TAC 356 Glu His Tyr Leu Gly Thr Glu Lys Leu Asp Ser Val Asn Ala Lys Tyr Leu Gly Arg Thr Ser Phe Asp Arg Asn Asn Trp Thr Leu Arg Leu His 50 AAT GTT CAG ATC AAG GAC ATG GGC TCG TAT GAT TGT TTT ATA CAA A~A 452 Asn Val Gln Ile Lys Asp Met Gly Ser Tyr Asp Cys Phe Ile Gln Lys ~WO 9~;/03408 ~ ~ 6 ~ 9 ~1 PCT/US94/08423 Lys Pro Pro Thr Gly Ser Ile Ile Leu Gln Gln Thr Leu Thr Glu ~eu Ser Val Ile Ala Asn Phe Ser Glu Pro Glu Ile Lys Leu Ala Gln Asn 0 Val Thr Gly Asn Ser Gly Ile Asn Leu Thr Cys Thr Ser Lys Gln Gly His Pro Lys Pro Lys Lys Met Tyr Phe Leu Ile Thr Asn Ser Thr Asn Glu Tyr Gly Asp Asn Met Gln Ile Ser Gln Asp Asn Val Thr Glu Leu Phe Ser Ile Ser Asn Ser Leu Ser Leu Ser Phe Pro Asp Gly Val Trp His Met Thr Val Val Cys Val Leu Glu Thr Glu Ser Met Lys Ile Ser 30 Ser Lys Pro Leu Asn Phe Thr Gln Glu Phe Pro Ser Pro Gln Thr Tyr Trp Lys Glu Ile Thr Ala Ser Val Thr Val Ala Leu Leu Leu Val Met Leu Leu Ile Ile Val Cys His Lys Lys Pro Asn Gln Pro Ser Arg Pro Ser Asn Thr Ala Ser Lys Leu Glu Arg Asp Ser Asn Ala Asp Arg Glu Thr Ile Asn Leu Lys GlU Leu Glu Pro Gln Ile Ala Ser Ala Lys Pro AAT GCA GAG TGAAGGCAGT GAGAGCCTGA GGA~AGAGTT AAAAATTGCT 1077 50 Asn Ala Glu (2) INFORMATION FOR SEQ ID NO:23:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 309 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) S~Qu~N~ DESCRIPTION: SEQ ID NO:23:
Met Asp Pro Arg Cys Thr Met Gly Leu Ala Ile Leu Ile Phe Val Thr Val Leu Leu Ile Ser Asp Ala Val Ser Val Glu Thr Gln Ala Tyr Phe 20 Asn Gly Thr Ala Tyr Leu Pro Cys Pro Phe Thr Lys Ala Gln Asn Ile Ser Leu Ser Glu Leu Val Val Phe Trp Gln Asp Gln Gln Lys Leu Val Leu Tyr Glu His Tyr Leu Gly Thr Glu Lys Leu Asp Ser Val Asn Ala Lys Tyr Leu Gly Arg Thr Ser Phe Asp Arg Asn Asn Trp Thr Leu Arg Leu His Asn Val Gln Ile Lys Asp Met Gly Ser Tyr Asp Cys Phe Ile Gln Lys Lys Pro Pro Thr Gly Ser Ile Ile Leu Gln Gln Thr Leu Thr Glu Leu Ser Val Ile Ala Asn Phe Ser Glu Pro Glu Ile Lys Leu Ala = 130 135 140 Gln Asn Val Thr Gly Asn Ser Gly Ile Asn Leu Thr Cys Thr Ser Lys Gln Gly His Pro Lys Pro Lys Lys Met Tyr Phe Leu Ile Thr Asn Ser Thr Asn Glu Tyr Gly Asp Asn Met Gln Ile Ser Gln Asp Asn Val Thr Glu Leu Phe Ser Ile Ser Asn Ser Leu Ser Leu Ser Phe Pro Asp Gly Val Trp His Met Thr Val Val Cys Val Leu Glu Thr Glu Ser Met Lys ~WO 95/03408 ~16 7 ~ 91 PCTIUS94/08423 Ile Ser Ser Lys Pro Leu Asn Phe Thr Gln Glu Phe Pro Ser Pro Gln Thr Tyr Trp Lys Glu Ile Thr Ala Ser Val Thr Val Ala Leu Leu Leu Val Met Leu Leu Ile Ile Val Cys His Lys Lys Pro Asn Gln Pro Ser Ary Pro Ser Asn Thr Ala Ser Lys Leu Glu Arg Asp Ser Asn Ala Asp Arg Glu Thr Ile Asn Leu Lys Glu Leu Glu Pro Gln Ile Ala Ser Ala Lys Pro Asn Ala Glu (2) INFORMATION FOR SEQ ID NO:24:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (c) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:24:

(2) INFORMATION FOR SEQ ID NO:25:
(i) ~QU~N~ CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRAN~N~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:25:

(2) INFORMATION FOR SEQ ID NO:26:

W O 9S/03408 PCTrUS94/08423 _ 2~709~ --(i) S~U~N~ CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: s ingle (D) TOPOLOGY: linear (ii) MOLECULE TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:26:

(2) INFORMATION FOR SEQ ID NO:27:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULB TYPE: oligonucleotide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:27:

(2) INFORMATION FOR SEQ ID NO:28:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1491 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA to mRNA
(iii) HYPOTHETICAL: no (iv) ANTI-SENSE: no (vi) ORIGINAL SOURCE:
(A) ORGANISM: HomQ sapi~n (F) TISSUE TYPE: lymphoid (G) CELL TYPE: B cell (H) CELL LINE: Raji ~WO 95/03408 2 ~ S ~ O 9 I PCTlUS94108423 (vii~ IMMEDIATE SOURCE:
(A) LIBRARY: cDNA in pCDM8 vector (B) CLONE: B7, Raji clone #13 (viii) POSITION IN GENOME:
(A) CHROMOSOME/SEGMENT: 3 (ix) FEATURE:
(A) NAME/KEY: Open reading frame (translated region) (B) LOCATION: 318 to 1181 bp (C) IDENTIFICATION METHOD: similarity to other pattern (ix) FEATURE:
(A) NAME/KEY: Alternate polyadenylation signal (B) LOCATION: 1474 to 1479 bp (C) IDENTIFICATION METHOD: similarity to other pattern (x) PUBLICATION INFORMATION:
(A) AUTHORS: FREEMAN, GORDON J.
FREEDMAN, ARNOLD S.
SEGIL, JEFFREY M.
LEE, GRACE
WHITMAN, JAMES F.
NADLER, LEE M.
(B) TITLE: B7, A New Member Of The Ig Superfamily With Unique Expression On Activated And Neoplastic B Cells (c) JOURNAL: The Journal of Immunology (D) VOLUME: 143 (E) ISSUE: 8 (F) PAGES: 2714-2722 (G) DATE: 15-OCT-1989 (H) RELEVANT RESIDUES In SEQ ID NO:28: FROM 1 TO 1491 (xi) ~Uu~ DESCRIPTION: SEQ ID NO:28:

GGAGTCTTAC CCTGAAATCA AAGGATTTAA AGAAAAAGTG GAALLlLl~l~ TCAGCAAGCT 120 GTGAAACTAA ATCCACAACC TTTGGAGACC CAGGAACACC CTCCAATCTC 'l'~'l'~'L~'l"L-l"l' 180 TTGCACCTGG GAAGTGCCCT GGTCTTACTT GGGTCCA~AT TGTTGGCTTT CACTTTTGAC 300 wo 95,03408 ~ ~ ~ 7 o ~ ~ PCT/US94/08423 ~

Met Gly His Thr Arg Arg Gln Gly Thr Ser Pro Ser S

Lys Cys Pro Tyr Leu Asn Phe Phe Gln Leu Leu Val Leu Ala Gly Leu TCT CAC TTC TGT TCA GGT GTT ATC CAC GTG ACC AAG GAA GTG A~A GAA 449 Ser His Phe Cys Ser Gly Val Ile His Val Thr Lys Glu Val Lys Glu Val Ala Thr Leu Ser Cys Gly His Asn Val Ser Val Glu Glu Leu Ala CAA ACT CGC ATC TAC TGG CAA AAG GAG AAG A~A ATG GTG CTG ACT ATG 545 Gln Thr Arg Ile Tyr Trp Gln Lys Glu Lys Lys Met Val Leu Thr Met Met Ser Gly Asp Met Asn Ile Trp Pro Glu Tyr Lys Asn Arg Thr Ile Phe Asp Ile Thr Asn Asn Leu Ser Ile Val Ile Leu Ala Leu Arg Pro TCT GAC GAG GGC ACA TAC GAG TGT GTT GTT CTG AAG TAT GAA A~A GAC 689 Ser Asp Glu Gly Thr Tyr Glu Cys Val Val Leu Lys Tyr Glu Lys Asp GCT TTC AAG CGG GAA CAC CTG GCT GAA GTG ACG TTA TCA GTC A~A GCT 737 Ala Phe Lys Arg Glu His Leu Ala Glu Val Thr Leu Ser Val Lys Ala Asp Phe Pro Thr Pro Ser Ile Ser Asp Phe Glu Ile Pro Thr Ser Asn Ile Arg Arg Ile Ile Cys Ser Thr Ser Gly Gly Phe Pro Glu Pro His ~1 67~91 ~WO 95/03408 PCT/US94/08423 Leu Ser Trp Leu Glu Asn Gly Glu Glu Leu Asn Ala Ile Asn Thr Thr Val Ser Gln Asp Pro Glu Thr Glu Leu Tyr Ala Val Ser Ser Lys Leu Asp Phe Asn Met Thr Thr Asn His Ser Phe Met Cys Leu Ile Lys Tyr Gly His Leu Arg Val Asn Gln Thr Phe Asn Trp Asn Thr Thr Lys Gln Glu His Phe Pro Asp Asn Leu Leu Pro Ser Trp Ala Ile Thr Leu Ile Ser Val Asn Gly Ile Phe Val Ile Cy8 Cy8 Leu Thr Tyr Cy8 Phe Ala Pro Ary Cy8 Arg Glu Arg Arg Arg Asn Glu Arg Leu Arg Arg Glu Ser Val Arg Pro Val TAATGTAACC l~llLlLllG CCATGTTTCC ATTCTGCCAT CTTGAATTGT ~ll~l~AGCC 1461 55 (3) INFORMATION FOR SEQ ID NO:29:

WO 95/03408 PCT/US94/08423 ~
~t~ 114-(i) SEQUENCE CHARACTERISTICS:
(A~ LENGTH: 288 amino acids (B) TYPE: amino acid (C) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (A) DESCRIPTION: B cell activation antigen; natural ligand for CD28 T cell surface antigen; tr~n~m~mhrane protein (ix) FEATURE:
(A) NAME/KEY: signal æequence (B) LOCATION: -34 to -1 (C) IDENTIFICATION METHOD: amino terminal sequencing of soluble protein (D) OTHER INFORMATION: hydrophobic (ix) FEATURE:
(A) NAME/KEY: extracellular domain (B) LOCATION: 1 to 208 (c) lV~NllrlCATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: tr~n~m~mhrane domain (B) LOCATION: 209 to 235 (C) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: intracellular domain (B) LOCATION: 236 to 254 (C) lv~Nll~lCATION METHOD: similarity with known sequence 45 ( ix) FEATURE:
(A) NAME/KEY: N-linked glycosylation (B) LOCATION: 19 to 21 (C) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: N-linked glycosylation ~O 95/03408 21 ~ 7 0 9 I PCT/US94/08423 (B) LOCATION: 55 to 57 (C) IDENTIFICATION METHOD: similarity with known æequence (ix) FEATURE:
(A) NAME/ Æ Y: N-linked glycosylation (B) LOCATION: 64 to 66 (C) lv~NLl~lCATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/ Æ Y: N-linked glycosylation (B) LOCATION: 152 to 154 (C) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/ Æ Y: N-linked glycosylation (B) LOCATION: 173 to 175 (C) l~Nll~lCATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: N-linked glycosylation (B) LOCATION: 177 to 179 (C) lv~Nll~lCATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/ Æ Y: N-linked glycosylation (B) LOCATION: 192 to 194 (C) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/ Æ Y: N-linked glycosylation (B) LOCATION: 198 to 200 (C) IDENTIFICATION METHOD: similarity with known sequence WO 95/03408 ~ ~ ~ 7 ~ 91 PCT/US94/084~3 tix) FEATURE:
(A) NAME/KEY: Ig V-set domain (B) LOCATION: 1 to 104 (c) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: Ig C-set domain (B) LOCATION: 105 to 202 (C) IDENTIFICATION METHOD: similarity with known sequence (x) PUBLICATION INFORMATION:
(A) AUTHORS: FREEMAN, GORDON J.
FREEDMAN, ARNOLD S.
SEGIL, JEFFREY M.
LEE, GRACE
WHITMAN, JAMES F.
NADLER, LEE M.
(B) TITLE: B7, A New Member Of The Ig Superfamily With Unique Expression On Activated And Neoplastic B Cells (C) JOURNAL: The Journal of Immunology (D) VOLUME: 143 (E) ISSUE: 8 (F) PAGES: 2714-2722 (G) DATE: 15-OCT-1989 (H) RELEVANT RESIDUES IN SEQUENCE ID NO:29, From -26 to 262 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29:

Met Gly His Thr Arg Arg Gln Gly Thr Ser Pro Ser Lys Cys Pro Tyr Leu Asn Phe Phe Gln Leu Leu Val Leu Ala Gly Leu Ser His Phe Cys Ser Gly Val Ile His Val Thr Lys Glu Val Lys Glu Val Ala Thr Leu Ser Cys Gly His Asn Val Ser Val Glu Glu Leu Ala Gln Thr Arg Ile 50 Tyr Trp Gln Lys Glu Lys Lys Met Val Leu Thr Met Met Ser Gly Asp Met Asn Ile Trp Pro Glu Tyr Lys Asn Arg Thr Ile Phe Asp Ile Thr 6~
~WO 95/03408 PCT/US94/08423 Asn Asn Leu Ser Ile Val Ile Leu Ala Leu Arg Pro Ser Asp Glu Gly Thr Tyr Glu Cys Val Val Leu Lys Tyr Glu Lys Asp Ala Phe Lys Arg 80 85 9o Glu His Leu Ala Glu Val Thr Leu Ser Val Lys Ala Asp Phe Pro Thr Pro Ser Ile Ser Asp Phe Glu Ile Pro Thr Ser Asn Ile Arg Arg Ile Ile Cys Ser Thr Ser Gly Gly Phe Pro Glu Pro His Leu Ser Trp Leu Glu Asn Gly Glu Glu Leu Asn Ala Ile Asn Thr Thr Val Ser Gln Asp Pro Glu Thr Glu Leu Tyr Ala Val Ser Ser Lys Leu Asp Phe Asn Met Thr Thr Asn His Ser Phe Met Cys Leu Ile Lys Tyr Gly His Leu Arg Val Asn Gln Thr Phe Asn Trp Asn Thr Thr Lys Gln Glu His Phe Pro Asp Asn Leu Leu Pro Ser Trp Ala Ile Thr Leu Ile Ser Val Asn Gly Ile Phe Val Ile Cys Cys Leu Thr Tyr Cys Phe Ala Pro Arg Cys Arg 35 Glu Arg Arg Arg Asn Glu Arg Leu Arg Arg Glu Ser Val Arg Pro Val 40 (4) INFORMATION FOR SEQ ID NO:30:
(i) SEQUENCE CHARACTBRISTICS:
(A) LENGTH: 1716 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULAR TYPE: cDNA to mRNA
- (iii) HYPOTHETICAL: no (vi) ORIGINAL SOURCE:
(A) ORGANISM: ~a m W 0 95/03408 PCTrUS94/08423 ~ a ~ 18-(D) DEVELOPMENTAL STAGE: germ line (F) TISSUE TYPE: lymphoid (G) CELL TYPE: B lymphocyte (H) CELL LINE: 70Z and A20 (vii) IMMEDIATE SOURCE:
(A) LIBRARY: cDNA in pCDM8 vector (B) CLONE: B7 #'s 1 and 29 (ix) FEATURE:
(A) NAME/KEY: translated region (B) LOCATION: 249 to 1166 bp (C) IDENTIFICATION METHOD: similarity to other pattern (ix) FEATURE:
(A) NAME/REY: Alternate ATG initiation codons (B) LOCATION: 225 to 227 and 270 to 272 (C) IDENTIFICATION METHOD: similarity to other pattern (Xi ) ~QU~'N~ DESCRIPTION: SEQ ID NO:30:

GAGTTTTATA CCTCAATAGA CTCTTACTAG ~ l TCAGGTTGTG AAACTCAACC 60 TTCAAAGACA ~l~l~lLCCA lll~l~lGGA CTAATAGGAT CATCTTTAGC ATCTGCCGGG 120 TGGATGCCAT CCAGGCTTCT llll~LACAT ~1~l~lll~l CGALllll~l~ GAGCCTAGGA 180 Met Ala Cys Asn Cys Gln Leu Met Gln Asp Thr Pro Leu Leu Lys Phe Pro Cys Pro Arg Leu Ile Leu Leu Phe Val Leu Leu Ile Arg Leu Ser Gln Val Ser Ser Asp Val Asp Glu Gln Leu Ser Lys Ser Val Lys Asp Lys Val Leu Leu Pro Cys Arg Tyr Asn Ser Pro His Glu Asp GAG TCT GAA GAC CGA ATC TAC TGG CAA A~A CAT GAC A~A GTG GTG CTG 482 Glu Ser Glu Asp Arg Ile Tyr Trp Gln Lys His Asp Lys Val Val Leu ~WO 95/03408 21 6 7 0 91 PCT/US94/08423 Ser Val Ile Ala Gly Lys Leu Lys Val Trp Pro Glu Tyr Lys Asn Arg Thr Leu Tyr Asp Asn Thr Thr Tyr Ser Leu Ile Ile Leu Gly Leu Val 0 Leu Ser Asp Arg Gly Thr Tyr Ser Cy8 Val Val Gln Lys Lys Glu Arg Gly Thr Tyr Glu Val Lys His Leu Ala Leu Val Lys Leu Ser Ile Lys Ala Asp Phe Ser Thr Pro Asn Ile Thr Glu Ser Gly Asn Pro Ser Ala ilO 115 120 Asp Thr Lys Arg Ile Thr Cys Phe Ala Ser Gly Gly Phe Pro Lys Pro Arg Phe Ser Trp Leu Glu Asn Gly Arg Glu Leu Pro Gly Ile Asn Thr Thr Ile Ser Gln Asp Pro Glu Ser Glu Leu Tyr Thr Ile Ser Ser Gln Leu Asp Phe Asn Thr Thr Arg Asn His Thr Ile Lys Cys Leu Ile Lys Tyr Gly Asp Ala His Val Ser Glu Asp Phe Thr Trp Glu Lys Pro Pro GAA GAC CCT CCT GAT AGC AAG AAC ACA CTT GTG CTC TTT GGG GCA GGA lOlo Glu Asp Pro Pro Asp Ser Lys Asn Thr Leu Val Leu Phe Gly Ala Gly Phe Gly Ala Val Ile Thr Val Val Val Ile Val Val Ile Ile Lys Cys 50 Phe Cys Lys His Arg Ser Cys Phe Arg Arg Asn Glu Ala Ser Arg Glu Thr Asn Asn Ser Leu Thr Phe Gly Pro Glu Glu Ala Leu Ala Glu Gln WO 9~/03408 PCT/US94/08423 ~
~1~7091 Thr Val Phe Leu ACAAGATAGA GTTAACTGGG AAGAGA~AGC CTTGAATGAG GATTTCTTTC CATCAGGAAG 1326 GCTGTCACTA A~AGGAGAGG TGCCTAGTCT TACTGCAACT TGATATGTCA TGTTTGGTTG 1506 15 GTGTCTGTGG GAGGCCTGCC ~Ll~ AA GAGAAGTGGT GGGAGAGTGG ATGGGGTGGG 1566 GTGGGGA~AA CTATGGTTGG GATGTA~AAA CGGATAATAA TATAAATATT A~ATAAAAAG 1686 AGAGTATTGA GC~AAAAA AAA~L~aAA 1716 (5) INFORMATION FOR SEQ ID NO:31:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 306 amino acids (B) TYPE: amino acid (c) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (A) DESCRIPTION: B lymphocyte activation antigen; Ig superfamily member; T cell costimulatory signal via activation of CD28 pathways, binds to CD28 T cells, tr~n~m~mhrane protein (ix) FEATURE:
(A) NAME/KEY: signal sequence (B) LOCATION: -37 to -1 (C) IDENTIFICATION METHOD: similarity with known sequence (D) OTHER INFORMATION: hydrophobic (ix) FEATURE:
(A) NAME/KEY: extracellular domain (B) LOCATION: 1 to 210 (C) l~NLl~lCATION METHOD: similarity with known sequence ~0 95/03408 ~ 1 6 ~ ~ ~1 PCT/US94/08423 (ix) FEATURE:
(A) NAME/KEY: transmembrane domain (B) LOCATION: 211 to 235 (C) l~Nll~lCATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: intracellular (cytoplasmic) domain (B) LOCATION: 236 to 269 (C) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: Ig V-set domain (B) LOCATION: 1 to 105 (c) IDENTIFICATION METHOD: similarity with known sequence (ix) FEATURE:
(A) NAME/KEY: Ig C-set domain (B) LOCATION: 106 to 199 (C) l~NLl~lCATION METHOD: similarity with known sequence (x) P B LICATION INFORMATION:
(A) AUTHORS: FREEMAN, GORDON J.
3 5 GRAY, GARY S.
GIMMI, CLAUDE D.
LOMBARD, DAVID B.
ZHOU, LIANG-JI
WHITE, MICHAEL
FINGEROTH, JOYCE D.
~RTRR~N, JOHN G.
NADLER, LEE M.
(B) TITLE: Structure, Expression, and T Cell Costimulatory Activity Of The Murine Homologue O~ The Human B
Lymphocyte Activation Antigen B7 (C) JOURNAL: Journal of Experimental Medicine (D) VOLUME:
(E) ISSUE:
(F) PAGES:
- (G) DATE: IN PRESS
(H) RELEVANT RESIDUES IN SEQUENCE ID NO:31: From -37 to 269 (xi) ~Qu~ DESCRIPTION: SEQ ID NO:31:

Met Ala Cys Asn Cys Gln Leu Met Gln Asp Thr Pro Leu Leu Lys Phe Pro Cys Pro Arg Leu Ile Leu Leu Phe Val Leu Leu Ile Arg Leu Ser Gln Val Ser Ser Asp Val Asp Glu Gln Leu Ser Lys Ser Val Lys Asp Ly~ Val Leu Leu Pro Cys Arg Tyr Asn Ser Pro His Glu Asp Glu Ser 15 Glu Asp Arg Ile Tyr Trp Gln Lys His Asp Lys Val Val Leu Ser Val Ile Ala Gly Lys Leu Lys Val Trp Pro Glu Tyr Lys Asn Arg Thr Leu Tyr Asp Asn Thr Thr Tyr Ser Leu Ile Ile Leu Gly Leu Val Leu Ser Asp Arg Gly Thr Tyr Ser Cys Val Val Gln Lys Lys Glu Arg Gly Thr Tyr Gly Val Lys His Leu Ala Leu Val Lys Leu Ser Ile Lys Ala Asp 30 Phe Ser Thr Pro Asn Ile Thr Glu Ser Gly Asn Pro Ser Ala Asp Thr Lys Arg Ile Thr Cys Phe Ala Ser Gly Gly Phe Pro Lys Pro Arg Phe Ser Trp Leu Glu Asn Gly Arg Glu Leu Pro Gly Ile Asn Thr Thr Ile 140 145 lS0 155 Ser Gln Asp Pro Glu Ser Glu Leu Tyr Thr Ile Ser Ser Gln Leu Asp Phe Asn Thr Thr Arg Asn His Thr Ile Lys Cys Leu Ile Lys Tyr Gly 5 Asp Ala His Val Ser Glu Asp Phe Thr Trp Glu Lys Pro Pro Glu Asp Pro Pro Asp Ser Lys Asn Thr Leu Val Leu Phe Gly Ala Gly Phe Gly Ala Val Ile Thr Val Val Val Ile Val Val Ile Ile Lys Cys Phe Cys Lys His Arg Ser Cys Phe Arg Arg Asn Glu Ala Ser Arg Glu Thr Asn ~VO 95/03408 2 t 6 ~ Q 91 PCT/US94/08423 Asn Ser Leu Thr Phe Gly Pro Glu Glu Ala Leu Ala Glu Gln Thr Val Phe Leu

Claims

1. An isolated nucleic acid comprising a nucleotide sequence encoding a peptide having an activity of a B lymphocyte antigen, B7-2.

2. The isolated nucleic acid of claim 1 which is a cDNA sequence.

3. The isolated nucleic acid of claim 2, wherein the cDNA is of human origin.

4. The isolated nucleic acid of claim 3, wherein the cDNA comprises a nucleotide sequence shown in Figure 8 (SEQ ID NO: 1).

5. The isolated nucleic acid of claim 3, wherein the cDNA comprises the coding region of a nucleotide sequence shown in Figure 8 (SEQ ID NO: 1).

6. The isolated nucleic acid of claim 2, wherein the cDNA is of murine origin.

7. The isolated nucleic acid of claim 6, wherein the cDNA comprises a nucleotide sequence shown in Figure 14 (SEQ ID NO:22).

8. The isolated nucleic acid of claim 6, wherein the cDNA comprises the coding region of a nucleotide sequence shown in Figure 14 (SEQ ID NO:22).

9. The isolated nucleic acid of claim 1, wherein the peptide comprises an amino acid sequence shown in Figure 8 (SEQ ID NO:2).

10. The isolated nucleic acid of claim 1, wherein the peptide comprises an aminoacid sequence shown in Figure 14 (SEQ ID NO:23).

11. The isolated nucleic acid of claim 1, wherein the peptide is at least 50%
homologous with a sequence comprising an amino acid sequence of Figure 8 (SEQ IDNO:2).

12. The isolated nucleic acid of claim 1, wherein the peptide is encoded by a nucleic acid which hybridizes under high or low stringency conditions to a nucleic acid which encodes a peptide comprising an amino acid sequence shown in Figure 8 (SEQ ID NO:2).

13. The isolated nucleic acid of claim 1, wherein the peptide is at least 20 amino acid residues in length.

14. The isolated nucleic acid of claim 1, wherein the peptide is at least 50%
homologous with a sequence comprising an amino acid sequence of Figure 14 (SEQ ID
NO:23).

15. The isolated nucleic acid of claim 1, wherein the peptide is encoded by a nucleic acid which hybridizes under high or low stringency conditions to a nucleic acid which encodes a peptide comprising an amino acid sequence shown in Figure 14 (SEQ ID NO:23).

16. The isolated nucleic acid of claim 15, wherein the peptide is at least 20 amino acid residues in length.

17. The isolated nucleic acid of claim 1, wherein the peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2).

18. An isolated DNA comprising a nucleotide sequence encoding a peptide having an activity of a B lymphocyte antigen, B7-2, the peptide having an amino acid sequence represented by formula Xn-Y-Zm, wherein Y comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2), wherein Xn is amino acid residues selected from amino acid residues contiguous to the amino terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2), wherein Zm is amino acid residues selected from amino acid residues contiguous to the carboxy terminus of Y in the sequence shown in Figure 8 (SEQ ID
NO:2), wherein n=0-23 and wherein m=0-84 .

19. The isolated DNA of claim 18, wherein n=0 and m=0.

20. The isolated DNA comprising a nucleotide sequence encoding a peptide of at least 20 amino acid residues or more in length and having at least about 50% homology with an amino acid sequence comprising a sequence shown in Figure 8 (SEQ ID NO:2).

21. An isolated nucleic acid encoding a B7-2 fusion protein comprising a nucleotide sequence encoding a first peptide having a B7-2 activity and a nucleotide sequence encoding a second peptide corresponding to a moiety that alters the solubility, binding affinity or valency of the first peptide.

22. The isolated nucleic acid of claim 21 which is a DNA.

23. The isolated nucleic acid of claim 22, wherein the first peptide comprises an extracellular domain of a human B7-2 protein.

24. The isolated nucleic acid of claim 23, wherein the first peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2).

25. The isolated nucleic acid of claim 23, wherein the first peptide comprises avariable region-like domain of human B7-2.

26. The isolated nucleic acid of claim 23, wherein the first peptide comprises aconstant region-like domain of human B7-2.

27. The isolated nucleic acid of claim 22, wherein the second peptide comprises an immonoglobulin constant region.

28. The isolated nucleic acid of claim 27, wherein the immunoglobulin constant region is a C.gamma.1 domain, including the hinge, CH2 and CH3 region.

29. The isolated nucleic acid of claim 27, wherein the immunoglobulin constant region is modified to reduce constant region-mediated biological effector functions.

30. The isolated nucleic acid of claim 29, wherein the biological effector function is selected from the group consisting of complement activation, Fc receptor interaction, and complement activation and Fc receptor interaction.

31. The isolated nucleic acid of claim 30, wherein the immunoglobulin constant region is a C.gamma.4 domain, including the hinge, CH2 and CH3 region.

32. The isolated nucleic acid of claim 31, wherein at least one amino acid residue of the CH2 domain is modified by substitution, addition or deletion.

33. An isolated B7-2 fusion protein comprising a first peptide having a B7-2 activity and a second peptide corresponding to a moiety that alters the solubility, binding affinity or valency of the first peptide.

34. The isolated B7-2 fusion protein of claim 33, wherein the first peptide comprises an extracellular domain of human B7-2 protein.

35. The isolated B7-2 fusion protein of claim 34, wherein the first peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2).

36. The isolated B7-2 fusion protein of claim 34, wherein the first peptide comprises a variable region-like domain of human B7-2.

37. The isolated B7-2 fusion protein of claim 34, wherein the first peptide comprises a constant region-like domain of human B7-2.

38. The isolated B7-2 fusion protein of claim 33, wherein the second peptide comprises an immonoglobulin constant region.

39. The isolated B7-2 fusion protein of claim 38, wherein the immunoglobulin constant region is a C.gamma.1 domain, including the hinge, CH2 and CH3 region.

40. The isolated B7-2 fusion protein of claim 38, wherein the immunoglobulin constant region is modified to reduce constant region-mediated biological effector functions.

41. The isolated B7-2 fusion protein of claim 40, wherein the biological effector function is selected from the group consisting of complement activation, Fc receptor interaction, and complement activation and Fc receptor interaction.

42. The isolated B7-2 fusion protein of claim 41, wherein the immunoglobulin constant region is a C.gamma.4 domain, including the hinge, CH2 and CH3 region.

43. The isolated B7-2 fusion protein of claim 42, wherein at least one amino acid residue of the CH2 domain is modified by substitution, addition or deletion.

44. A composition suitable for pharmaceutical administration comprising a fusion protein of claim 33 and a pharmaceutically acceptable carrier.

45. A composition suitable for pharmaceutical administration comprising a fusion protein of claim 34 and a pharmaceutically acceptable carrier.

46. A composition suitable for pharmaceutical administration comprising a fusionprotein of claim 36 and a pharmaceutically acceptable carrier.

47. A composition suitable for pharmaceutical administration comprising a fusionprotein of claim 38 and a pharmaceutically acceptable carrier.

48. A recombinant expression vector comprising a nucleic acid of claim 1.

49. The recombinant expression vector of claim 48, wherein the nucleic acid is acDNA sequence.

50. The recombinant expression vector of claim 49, wherein the cDNA is of human origin and comprises a nucleotide sequence shown in Figure 8 (SEQ ID NO:1).

51. The recombinant expression vector of claim 49 which is a plasmid.

52. A recombinant expression vector comprising a nucleic acid of claim 7.

53. A host cell transfected with the expression vector of claim 48 capable of directing the expression of a peptide having an activity of a B lymphocyte antigen, B7-2.

54. A host cell transfected with the expression vector of claim 50 capable of directing the expression of a peptide having an activity of a B lymphocyte antigen, B7-2.

55. A host cell transfected with the expression vector of claim 52 capable of directing the expression of a peptide having an activity of a B lymphocyte antigen, B7-2.

56. An isolated, recombinant peptide having an activity of a B lymphocyte antigen, B7-2, expressed by a host cell of claim 54.

57. A cell transfected with a nucleic acid encoding a peptide having an activity of a B lymphocyte antigen, B7-2, in a form suitable for expression of the peptide on the cell surface.

58. The cell of claim 57, wherein the nucleic acid is a cDNA comprising a nucleotide sequence shown in Figure 8 (SEQ ID NO:1) in a recombinant expression vector.

59. A tumor cell which is modified to express a T cell costimulatory molecule,B7-2.

60. The tumor cell of claim 59 which is transfected with a nucleic acid encoding human B7-2 in a form suitable for expression of B7-2.

61. The tumor cell of claim 59 which is stimulated to express B7-2.

62. The tumor cell of claim 59 which has a human B7-2 antigen coupled to the tumor cell.

63. The tumor cell of claim 59 which expresses a T cell costimulatory molecule, B7-1.

64. The tumor cell of claim 59 which expresses a T cell costimulatory molecule, B7-3.

65. The tumor cell of claim 59 which expresses an MHC class I molecule.

66. The tumor cell of claim 59 which expresses an MHC class II molecule.

67. The tumor cell of claim 59 which normally expresses an MHC class II
associated protein, the invariant chain, and wherein expression of the invariant chain is inhibited.

68. A tumor cell transfected with a nucleic acid encoding a T cell costimulatory molecule, B7-2, in a form suitable for expression of B7-2.

69. The tumor cell of claim 68, wherein the nucleic acid is a cDNA in a recombinant expression vector.

70. The tumor cell of claim 68, further transfected with a nucleic acid encoding a T cell costimulatory molecule, B7-1, in a form suitable for expression of B7-1.

71. The tumor cell of claim 68, further transfected with a nucleic acid encoding a T cell costimulatory molecule, B7-3, in a form suitable for expression of B7-3.

72. The tumor cell of claim 68, further transfected with at least one nucleic acid comprising DNA encoding:
(a) at least one MHC class II .alpha. chain protein; and (b) at least one MHC class II .beta. chain protein, wherein the nucleic acid is in a form suitable for expression of the MHC class II .alpha. chain protein(s) and the MHC class II .beta. chain protein(s).

73. The tumor cell of claim 72 which does not express MHC class II molecules prior to transfection of the tumor cell.

74. The tumor cell of claim 68, further transfected with at least one nucleic acid encoding at least one MHC class I a chain protein in a form suitable for expression of the MHC class I protein(s).

75. The tumor of claim 74, further transfected with a nucleic acid encoding a .beta.-2 microglobulin protein in a form suitable for expression of the .beta.-2 microglobulin protein.

76. The tumor cell of claim 68 which normally expresses an MHC class II
associated protein, the invariant chain, and wherein expression of the invariant chain is inhibited.

77. The tumor cell of claim 76, wherein expression of the invariant chain is inhibited by transfection of the tumor cell with a nucleic acid which is antisense to a regulatory or a coding region of the invariant chain gene.

78. The tumor cell of claim 68 which is a sarcoma.

79. The tumor cell of claim 68 which is a lymphoma.

80. The tumor cell of claim 68 which is selected from a group consisting of a melanoma, a neuroblastoma, a leukemia and a carcinoma.

81. A method of treating a subject with a tumor, comprising:
(a) obtaining tumor cells from the subject;
(b) transfecting the tumor cells with a nucleic acid encoding B7-2 in a form suitable for expression of B7-2; and (c) administering the tumor cells to the subject.

82. The method of claim 81, wherein the tumor cells are further transfected with a nucleic acid encoding B7-1.

83. The method of claim 81, wherein the tumor cells are further transfected with at least one nucleic acid encoding at least one MHC class II .alpha. chain protein and at least one MHC class II .beta. chain protein in a form suitable for expression of the MHC class II .alpha. chain protein(s) and the MHC class II .beta. chain protein(s).

84. The method of claim 81, wherein the tumor cells are further transfected with at least one nucleic acid encoding at least one MHC class I .alpha. chain protein in a form suitable for expression of the MHC class I protein(s).

85. The method of claim 84, wherein the tumor cells are further transfected with a nucleic acid encoding a .beta.-2 microglobulin protein in a form suitable for expression of the .beta.-2 microglobulin protein.

86. The method of claim 81, wherein expression of an MHC class II associated protein, the invariant chain, is inhibited in the tumor cells.

87. The method of claim 86, wherein expression of the invariant chain is inhibited in the tumor cells by transfection of the tumor cell with a nucleic acid which is antisense to a regulatory or a coding region of the invariant chain gene.

88. The method of claim 81, wherein the tumor is a sarcoma.

89. The method of claim 81, wherein the tumor is a lymphoma.

90. The method of claim 81, wherein the tumor is selected from a group consisting of a melanoma, a neuroblastoma, a leukemia and a carcinoma.

91. A method of inducing an anti-tumor response by CD4+ T lymphocytes in a subject with a tumor, comprising:
(a) obtaining tumor cells from the subject;
(b) transfecting the tumor cells with at least one nucleic acid comprising DNA encoding:
(i) B7-2, (ii) an MHC class II .alpha. chain protein, and (iii) an MHC class II .beta. chain protein, wherein the nucleic acid is in a form suitable for expression of B7-2, the MHC class II .alpha.
chain protein and the MHC class II .beta. chain protein; and (c) administering the tumor cells to the subject.

92. A method for treating a subject with a tumor comprising modifying tumor cells in vivo to express a T cell costimulatory molecule, B7-2.

93. The method of claim 92, wherein tumor cells are modified in vivo by delivering to the subject in vivo a nucleic acid encoding B7-2 in a form suitable for expression of B7-2.

94. The method of claim 93, wherein the nucleic acid is delivered to the subject in vivo by injection of the nucleic acid in an appropriate vehicle into the tumor.

95. A method for treating a subject with a tumor, comprising:
(a) obtaining tumor cells and T lymphocytes from the subject;
(b) culturing the T lymphocytes from the subject in vitro with the tumor cells from the subject and with a stimulatory form of B7-2; and (c) administering the T lymphocytes to the subject.

96. A peptide having an activity of a B lymphocyte antigen, B7-2, produced by recombinant expression of a nucleic acid of claim 1.

97. A peptide having an activity of a B lymphocyte antigen, B7-2, produced by recombinant expression of a nucleic acid of claim 4.

98. A peptide having an activity of a B lymphocyte antigen, B7-2, produced by recombinant expression of a nucleic acid of claim 5.

99. A peptide of claim 98 comprising an amino acid sequence set forth in Figure 8 (SEQ ID NO: 2).

100. A peptide having an activity of a B lymphocyte antigen, B7-2, produced by recombinant expression of a DNA of claim 18.

101. A peptide having an activity of a B lymphocyte antigen, B7-2, produced by recombinant expression of a DNA of claim 20.

102. A substantially pure preparation of a peptide having an activity of a B
lymphocyte antigen, B7-2.

103. A substantially pure preparation of a peptide having an activity of a B
lymphocyte antigen, B7-3.

104. A peptide having an amino acid sequence represented by a formula Xn-Y-Zm, wherein Y is amino acid residues selected from the group consisting of: amino acid residues 55-68 of the sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 81-89 of the sequence shown in Figure 8 (SEQ ID NO: 2); amino acid residues 128-142 of the sequence shown in Figure 8 (SEQ ID NO: 2); amino acid residues 160-169 of the sequence shown in Figure 8 (SEQ ID NO:2); amino acid residues 188-200 of the sequence shown in Figure 8 (SEQ ID NO: 2), and amino acid residues 269-282 of the sequence shown in Figure 8 (SEQ
ID NO: 2), wherein Xn is amino acid residues selected from amino acid residues contiguous to the amino terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2), wherein Zm is amino acid residues selected from amino acid residues contiguous to the carboxy terminus of Y in the sequence shown in Figure 8 (SEQ ID NO:2), wherein n=0-30 and wherein m=0-30.

105. A peptide of claim 104, wherein n=0 and m=0.

106. An antibody specifically reactive with a peptide produced by recombinant expression of a nucleotide sequence encoding a peptide having an activity of a human B
lymphocyte antigen, B7-2.

107. The antibody of claim 106, wherein the nucleotide sequence comprises a coding region of a nucleotide sequence shown in Figure 8 (SEQ ID NO: 1).

108. The antibody of claim 106 which is a monoclonal antibody.

109. The antibody of claim 108 which is an IgG1 antibody.

110. The antibody of claim 108 which is an IgG2a antibody.

111. A hybridoma HF2.3D1 designated by ATCC Accession No.?.

112. A monoclonal antibody produced by the hybridoma of claim 111.

113. A hybridoma HA5.2B7 designated by ATCC Accession No.? .

114. A monoclonal antibody produced by the hybridoma of claim 113.

115. A hybridoma HA3.1F9 designated by ATCC Accession No.?.

116. A monoclonal antibody produced by the hybridoma of claim 115.

117. A nonhuman, transgenic animal which contains cells transfected to express apeptide having an activity of a B lymphocyte antigen, B7-2.

118. The nonhuman, transgenic animal of claim 117 which is a mouse.

119. A nonhuman, knockout animal which contains cells having an altered gene encoding a B lymphocyte antigen, B7-2.

120. The nonhuman, knockout animal of claim 119 which is a mouse.

121. A composition suitable for pharmaceutical administration comprising a peptide having an activity of a B lymphocyte antigen, B7-2, and a pharmaceutically acceptable carrier.

122. The composition of claim 121 further comprising a peptide having an activity of a B lymphocyte antigen, B7-1.

123. The composition of claim 121, wherein the peptide comprises an amino acid sequence set forth in Figure 8 (SEQ ID NO: 2).

124. The composition of claim 123, wherein the peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO: 2).

125. A method for producing a peptide having an activity of a B lymphocyte antigen, B7-2, comprising culturing a host cell of claim 53 in a medium to express the peptide and isolating the peptide from the medium.

126. A method for producing a peptide having an activity of a B lymphocyte antigen, B7-2, comprising culturing a host cell of claim 54 in a medium to express the peptide and isolating the peptide from the medium.

127. A method for inhibiting an interaction of a B lymphocyte antigen, B7-2, with its natural ligand(s) on the surface of immune cells, comprising contacting an immune cell with a reagent which inhibits B7-2 binding with its natural ligand(s), to thereby inhibit costimulation of the immune cell through the B7-2-ligand interaction.

178. The method of claim 126, wherein the reagent is a peptide having B7-2 binding activity, but lacking the ability to deliver a costimulatory signal to the immune cell.

129. The method of claim 128, wherein the peptide is a soluble, monomeric peptide.

130. The method of claim 129, wherein the peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO: 2).

131. The method of claim 130, wherein the reagent is a B7-2 fusion protein comprising a first peptide having B7-2 activity and a second peptide comprising a moiety that alters the solubility, binding affinity or valency of the first peptide.

132. The method of claim 131, wherein the first peptide comprises an extracellular domain of the human B7-2 protein.

133. The method of claim 132, wherein the first peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2).

134. The method of claim 131, wherein the second peptide comprises an immonoglobulin constant region.

135. The method of claim 134, wherein the immunoglobulin constant region is a C.gamma.
1 domain, including the hinge, CH2 and CH3 region.

136. The method of claim 131, wherein the reagent is an antibody reactive with B7-2.

137. The method of claim 136, wherein the antibody is a monoclonal antibody.

138. A method for downregulating T cell mediated immune responses in a subject, comprising administering to the subject an agent having B7-2 binding activity, but lacking the ability to deliver a costimulatory signal to T cells, in an amount effective to inhibit T cell proliferation and/or cytokine secretion in the subject.

139. The method of claim 138, wherein the agent is a peptide having B7-2 bindingactivity, but lacking the ability to deliver a costimulatory signal to T cells.

140. The method of claim 138, wherein the agent is an antibody reactive with B7-2.

141. The method of claim 140, wherein the antibody is a monoclonal antibody.

142. The method of claim 138, further comprising administering to the subject anagent having B7-1 binding activity, but lacking the ability to deliver a costimulatory signal to T cells.

143. The method of claim 142, wherein the agent is a peptide having B7-1 bindingactivity, but lacking the ability to deliver a costimulatory signal to T cells.

144. The method of claim 142, wherein the agent is an antibody reactive with B7-1.

145. The method of claim 144, wherein the antibody is a monoclonal antibody.

146. The method of claim 138, further comprising administering to the subject animmunomodulating reagent selected from the group consisting of an antibody reactive with CD28, an antibody reactive with CTLA4, an antibody reactive with a cytokine, a CTLA4Ig fusion protein, a CD28Ig fusion protein, and an immunosuppressive drug.

147. A method for treating an autoimmune disease in a subject mediated by interaction of a B lymphocyte antigen, B7-2, with its natural ligand(s) on the surface of immune cells, comprising administering to the subject an inhibitory form of B7-2 protein, to thereby inhibit costimulation of the immune cells through the B7-2-ligand interaction.

148. The method of claim 147, wherein the autoimmune disease is selected from the group consisting of diabetes mellitus, rheumatoid arthritis, multiple sclerosis, myasthenia gravis, systemic lupus enthmatosis, and autoimmune thyroiditis.

149. The method of claim 147, wherein the inhibitory form of B7-2 protein is a peptide having B7-2 binding activity, but lacking the ability to deliver a costimulatory signal to immune cells.

150. The method of claim 149, wherein the peptide is a soluble, monomeric peptide.

151. The method of claim 150, wherein the peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO: 2).

152. The method of claim 147, wherein the inhibitory form of B7-2 protein is a B7-2 immunoglobulin fusion protein (B7-2Ig) comprising a first peptide comprising an extracellular domain of the B7-2 protein and a second peptide comprising an immunoglobulin constant domain.

153. The method of claim 152, wherein the extracellular domain of the B7-2 protein comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID
NO:2).

154. The method of claim 147, wherein the inhibitory form of B7-2 protein is an antibody reactive with B7-2.

155. The method of claim 154, wherein the antibody is a monoclonal antibody.

156. The method of claim 149, further comprising administering to the subject a peptide having B7-1 binding activity, but lacking the ability to deliver a costimulatory signal to T cells.

157. The method of claim 147, further comprising administering to the subject animmunomodulating reagent selected from the group consisting of an antibody reactive with B7-1, an antibody reactive with CD28, an antibody reactive with CTLA4, an antibody reactive with a cytokine, a CTLA4Ig fusion protein, a CD28Ig fusion protein. and an immunosuppressive drug.

158. A method for treating allergy in a subject mediated by interaction of a B
lymphocyte antigen, B7-2, with its natural ligand(s) on the surface of immune cells, comprising adimistering to the subject an inhibitory form of B7-2 protein, to thereby inhibit costimulation of the immune cells through the B7-2 -ligand interaction.

159. The method of claim 158, wherein the inhibitory form of B7-2 protein is a peptide having B7-2 binding activity, but lacking the ability to deliver a costimulatory signal to immune cells.

160. A method for inhibiting donor T cell proliferation and/or cytokine secretion in a transplant recipient to thereby prevent graft-versus-host disease (GVHD) in the recipient, comprising contacting donor T cells to be transplanted with an agent having B7-2 binding activity, but lacking the ability to deliver a costimulatory signal to the T cells.

161. The method of claim 160, wherein the agent is a peptide having B7-2 bindingactivity, but lacking the ability to deliver a costimulatory signal to T cells.

162. The method of claim 161, wherein the peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO: 2).

163. The method of claim 160, wherein the agent is an antibody reactive with B7-2.

164. The method of claim 163, wherein the antibody is a monoclonal antibody.

165. A method for inhibiting transplantation rejection in a recipient of a tissue or organ transplant, comprising administering to the recipient an agent having B7-2 binding activity, but lacking the ability to deliver a costimulatory signal to the T cells.

166. The method of claim 165, wherein the agent is a peptide having B7-2 bindingactivity, but lacking the ability to deliver a costimulatory signal to T cells.

167. The method of claim 166, wherein the peptide comprises amino acid residues 24-245 of the sequence shown in Figure 8 (SEQ ID NO:2).

168. The method of claim 165, wherein the agent is an antibody reactive with B7-2.

169. The method of claim 168, wherein the antibody is a monoclonal antibody.

170. A method for upregulating T cell mediated immune responses in a subject, comprising administering to the subject a peptide having B7-2 activity, in an amount effective to stimulate T cell proliferation and/or cytokine secretion in the subject.

171. The method of claim 170, further comprising administering to the subject a peptide having B7-1 activity.

172. The method of claim 170, further comprising administering to the subject a pathogen or portion thereof to thereby induce an anti-pathogen immune response in the subject.

173. The method of claim 172, wherein the pathogen is a virus.

174. A method of identifying molecules which modulate expression of a B7-2 antigen, comprising a) contacting a cell which expresses a peptide having B7-2 activity with a molecule to be tested, under conditions appropriate for interaction of the molecule with the cell; and b) determining the effect of the molecule on cell expression of the peptide having B7-2 activity.

175. The method of claim 174, wherein the effect of the molecule on cell expression of the peptide having B7-2 activity is determined by detecting the presence of the peptide on the cell surface.

176. The method of claim 175, wherein the presence of the peptide on the cell surface is detected by immunofluorescence with an antibody reactive with the peptide or with a CTLA4Ig or CD28Ig fusion protein.

177. The method of claim 174, wherein the effect of the molecule on cell expression of the peptide having B7-2 activity is determined by detecting the presence of mRNA encoding the peptide in the cell.

178. The method of claim 177, wherein the presence of mRNA is detected by hybidization with B7-2 cDNA.

179. A method of identifying a cytokine produced by an immune cell in response to costimulation with a B7-2 antigen, comprising a) contacting an activated immune cell and a cell which expresses a peptide having B7-2 activity, in an appropriate cell culture medium; and b) determining the presence of a cytokine in the cell culture medium.

180. The method of claim 179, wherein the immune cell is a T cell.

181. The method of claim 179, wherein the presence of a cytokine in the cell culture medium is determined by contacting the medium with an antibody reactive with the cytokine.

182. A method of identifying molecules which inhibit costimulation of immune cells by a B7-2 antigen, comprising a) contacting an immune cell which has received a primary activation signal with a stimulatory form of B7-2 protein and a molecule to be tested, under conditions appropriate for interaction of the molecule with the immune cell and the stimulatory form of B7-2 protein; and b) determining the effect of the molecule on costimulation of the immune cell by the stimulatory form of B7-2 protein.

183. The method of claim 182, wherein the immune cell is a T cell.

184. The method of claim 183, wherein the effect of the molecule on costimulation of the T cell is determined by detecting T cell proliferation and/or cytokine production.

185. The method of claim 182, wherein the stimulatory form of B7-2 is a cell which expresses a peptide having B7-2 activity on the cell surface.

186. A method of identifying molecules which inhibit binding of a B7-2 antigen to a ligand on the surface of immune cells, comprising a) contacting a labeled B7-2 ligand and a molecule to be tested with a peptide having B7-2 activity;
b) removing unbound labeled B7-2 ligand; and c) determining the amount of labeled B7-2 ligand bound to the peptide having B7-2 activity, as an indication of the ability of the molecule to inhibit binding of the B7-2 ligand to a B7-2 antigen.

187. The method of claim 186, wherein the immune cell is a T cell and the B7-2 ligand is CTLA4 or CD28.

188. The method of claim 186, wherein the peptide is immobilized on a solid phase support.

189. A method of identifying molecules which inhibit intracellular signaling by an immune cell in response to a stimulatory form of a B7-2 protein, comprising a) contacting an immune cell which has received a primary activation signal and which expresses a B7-2 ligand on the cell surface with a stimulatory form of B7-2 protein and a molecule to be tested, under conditions appropriate for interaction of the molecule with the immune cell and the stimulatory form of B7-2 protein; and b) determining the effect of the molecule on intracellular signaling by the immune cell in response to the stimulatory form of B7-2 protein.

190. The method of claim 189, wherein the immune cell is a T cell.

191. The method of claim 190, wherein the effect of the molecule on intracellular signaling by the immune cell is determined by detecting T cell proliferation and/or cytokine production.

192. The method of claim 189, wherein the stimulatory form of B7-2 is a cell which expresses a peptide having B7-2 activity on the cell surface.

193. A method of isolating a B lymphocyte antigen, B7-3, comprising contacting a cell material which contains a peptide having B7-3 activity, with an antibody reactive with B7-3 under conditions appropriate for binding of the antibody to the peptide and isolating the peptide from the antibody.

194. The method of claim 193, wherein the antibody is a monoclonal antibody BB-1.