Title:  Antibody specific for H4-1BB

United States Patent:  6,569,997

Issued:  May 27, 2003

Inventors:  Kwon; Byoung S. (Carmel, IN)

Assignee:  Advanced Research and Technology Institute, Inc. (Bloomington, IN)

Appl. No.:  578764

Filed:  May 25, 2000

Disclosed herein are the methods of using the H4-1BB protein, ligands to this protein, and various mAbs either directed against H4-1BB or other molecules that can be used therapeutically. The nature and importance of the H4-1BB molecule provides the ligands and related co-stimulatory molecules the ability to enhance or suppress T-cell activation and proliferation. By treating T-cells that have expressed receptor protein H4-1BB with one of the four anti-H4-1BB monoclonal antibodies disclosed herein activation or inhibition of the immune response is seen. Also disclosed herein is cDNA for the human receptor H4-1BB. The cDNA of the human receptor H4-1BB is about 65% homologous to the mouse cDNA 4-1BB and was isolated by using probes derived from murine cDNA 4-1BB. A fusion protein for detecting cell membrane ligands to human receptor protein H4-1BB was developed. It comprises the extracellular portion of the receptor protein H4-1BB and a detection protein, alkaline phosphatase, bound to the portion of the receptor protein H4-1BB. B-cells that have expressed a ligand to receptor protein H4-1BB can be treated with cells that have expressed receptor protein H4-1BB and B-cell proliferation may be induced. The use of H4-1BB to block H4-1BB ligand binding has practical application in the suppression of the immune system during organ transplantation or against autoimmune diseases including diabetes, rheumatoid arthritis, and lupus. Other applications of this technology include the development of therapeutic methods for the treatment of HIV-1 infected individuals, and the treatment of cancerous tumors.

DETAILED DESCRIPTION OF THE INVENTION

The following description teaches the isolation of 4-1BB and its human homologue, H4-1BB, the preparation of the peripheral blood cells, including the antibodies and reagents used, the production of fusion protein, immunization, and the production of monoclonal antibodies acting as either agonists or antagonists of H4-1BB. Also disclosed is the therapeutic use of: 4-1BB, antibodies to it, its ligands, and immunoprecipitation studies.

Isolation and Characterization of Mouse Receptor 4-1BB

U.S. application Ser. No. 08/012,796 discloses the nucleotide sequence and the deduced amino acid sequence of the mouse receptor 4-1BB. The transcript of 4-1BB was inducible by concanavalin A in mouse splenocytes, T cell clones, and hybridomas. The expression of 4-1BB transcripts was inhibited by cyclosporin A. The 4-1BB mRNA was inducible by antigen receptor stimulation but was not inducible by IL-2 stimulation in the cloned T-cells (1). The 4-1BB cDNA encodes a peptide of 256 amino acids containing a putative leader sequence, a potential membrane anchor segment, and other features of known receptor proteins. Therefore, the expression pattern of 4-1BB resembles those of lymphokine mRNAs while the sequence appeared consistent with those of receptor proteins.

The major species of 4-1BB on the cell surface appears to be a 55-kDa dimer. 4-1BB also appears to exist as a 30-kDa monomer and possibly as a 110-kDa tetramer. Since these 4-1BB species were immunoprecipitated from a homogeneous population of cells (T-cell clone fl), all forms potentially co-exist on each cell. Peptide digests from the 4-1BB monomer and dimer are needed to determine whether 4-1BB exists as a homodimer on the cell surface. A variety of cell surface receptors such as the insulin receptor (Ebina et al., 1985), the B cell surface immunoglobin receptor (Vassali et al.,), the T cell Ag receptor skins et al., 1983), the CD-28 co-stimulatory receptor (Lesslaver et al., 1986), and the CD27 Tell antigen (Van Lier et al., 1987) are composed of disulfide-bonded subunits. Receptor dimerization may be required for ligand binding and subsequent cell signal transduction.

4-1BB is not expressed on resting T cells but is inducible by activators which deliver a complete growth stimulus to the T cell. The combination of phorbol myristate acetate (PMA) and ionomycin is capable of mimicking those signals required for T cell proliferation. Although PMA or ionomycin alone induced 4-1BB mRNA, the combination of PMA and ionomycin resulted in optimal 4-1BB expression. Furthermore, the expression of 4-1BB was not transient. When purified splenic T-cells were stimulated with immobilized anti-CD3, 4-1BB mRNA was expressed and this expression was maintained for up to 96 hrs post-stimulation. Cell cycle analysis will be required to confirm that 4-1BB is expressed throughout cell cycle progression.

4-1BB is structurally related to members of the nerve growth factor receptor super-family. Although these receptors possess structurally similar ligand-binding properties (cysteine-rich regions), the cytoplasmic domains of these proteins are non-conserved which could allow for diversity in transmembrane signaling. Some members of this family are involved in the T or B cell activation process. There are in vitro functional data on the OX-40, CD40 and CD27 antigens. Antibodies against the OX-40 augment the T-cell response in a mixed lymphocyte reaction and antibodies against CD40 enhance B-cell proliferation in the presence of a co-activator, such as PMA or CD20 antibodies, and synergizes with IL-4 in vitro to induce B-cell differentiation and to generate long-term normal B cell lines. One monoclonal antibody, anti-1A4, which recognizes an epitope on the CD27 molecule inhibited calcium mobilization, IL-2 secretion, helper T cell function, and T-cell proliferation. On the other hand, CLB-CD27/1, another anti-CD27 mAb enhanced proliferation of human T cells stimulated with phytohemagglutinin (PHA) or anti-CD3 mAb. These results indicate that the CD27 molecule plays an important role in T cell activation. Except for TNFR's, NCFR and CD40, the ligands or cell surface molecules to which the members of the superfamily bind are not yet identified. Identification and characterization of the ligands to which the receptors bind will be helpful in better defining the physiologic role of 4-1BB.

To ascertain whether cell surface 4-1BB could contribute to T cell activation, the anti-4-1BB 53A2 was used as an antagonist to 4-1BB. The resulting data suggest that 4-1BB does in fact have the potential to function as an accessory signaling molecule during T cell activation and proliferation. The addition of soluble 53A2 to purified splenic T cells stimulated with immobilized anti-CD3 resulted in an amplification of ³ H thymidine incorporation compared to T cells stimulated with anti-CD3 alone. This pattern of enhancement ranged from 2- to 10-fold in three independent experiments.

In the original two signal model of Bretcher and Cohn, they proposed that signal 1, the occupancy of the T cell antigen receptor (TCR), resulted in inactivation of the T cell in the absence of signal 2, which is provided by accessory cells. This has since been confirmed by a variety of studies (Moeller et al., 1989). The identification of the accessory cell CD28 as a potent co-stimulatory receptor on T cells was a significant contribution in beginning to characterize the accessory signal(s) required for optimal T cell proliferation. It is possible that other cell surface molecules may contribute to these co-stimulatory activation requirements.

The biochemical signals delivered through 4-1BB indicate that there is a putative p56^lck tyrosine kinase-binding domain in its cytoplasmic tail. It was later determined that p56^lck tyrosinase kinase binds to 4-1BB. It will also be worthwhile to determine if 4-1BB-mediated signaling can regulate genes such as IL-2 and IL-2 receptor, whose expression is required for T cell activation and subsequent proliferation.

The precise functions of members of the Nerve Growth Factor Receptor (NGFR) superfamily appear to be diverse. An emerging theme of inquiry concerns the ability of these molecules to maintain the responsiveness or viability of the particular cell type in which they are expressed. For instance, NGF is absolutely required for viability of neurons in vitro and in vivo (Yamori et al., 1992). The cross-linking of CD40 by soluble antiCD40 monoclonal antibody blocks germinal center centrocytes from undergoing apoptosis in vitro. Signals delivered through CD40 may also aid in maintenance of responsiveness to differentiation factors. The ligation of CD40 with anti-CD40 F(ab')₂ fragments in the presence of IL-4 induced large increases IgE synthesis, Also, anti-CD40 activated naive B cells treated with IL-10 and transforming growth factor-.beta. became committed to IgA secretion (DeFrance et al., 1992). In addition to sharing the molecular characteristics with the NGFR superfamily, it was noted that the 4-1BB contained a putative zinc finger structure similar to that of the yeast elF-2b protein. 4-1BB also shares a conserved region with the sina seven in absentia of Drosophila Melanogaster, which is required for correct photoreceptor cell development (Carthew and Rubin, 1990). That particular region is also similar to the protein product of the DG17 gene of Dictyostelium, whose expression is specifically induced during aggregation by cyclic adenosine monophosphate (cAMP).

This region forms the pattern of C--X₂ --C--X₉ --C--X₃ --H--X₃ --C--X--C (SEQ ID NO:9); and the cysteines and histidine are conserved in a similar space in 4-1BB, sina, and DG17 proteins. Ten of 24 amino acids between the 4-1BB and sina proteins are identical; 3 of 24 are conservative substitutes. The conserved pattern suggests that these amino acids are functionally important. The sina protein is localized in the nucleus, suggesting that it has a regulatory function in cells. The fact that the amino acid sequence of 4-1BB contains features like a zinc finger motif, a nuclear protein, and a receptor domain suggests that 4-1BB may play diverse roles during cellular proliferation and differentiation.

In addition, 4-1BB may represent another cell-surface molecule involved in T cell-antigen presenting cell (APC) interactions. The 4-1BB-alkaline phosphatase (4-1BB-AP) fusion protein specifically bound to mature lines, anti-IF-activated primary B cells, and mature macrophage-cell lines. 4-1BB-AP bound at low or insignificant levels to immature B- and macrophage-cell lines, T cell clones, T cell lines, primary culture T cells, and various non-lymphoid-cell lines. Since 4-1BB-AP binds to mature B cells and macrophages, it is possible that signals delivered upon 4-1BB binding may modulate APC functions in some way. This possibility remains to be explored.

Chalupny et al., proposed that 4-1BB Rg, a fusion protein consisting of the extracellular domain of 4-1BB and the Fc region of human IgG, bound to the extracellular matrix (ECM). The highest level of 4-1BB Rg binding was to human vitronectin. The inventors performed an ELISA to test this possibility using 4-1BB-AP and human vitronectin (Yelios Pharmaceuticals/GIBCO-BRL, Grand Island, N.Y.) immobilized at 0.007 mg, with 10 mg per well on microtiter plates. No binding of 4-1BB-AP based on AP activity was observed. To rule out the possibility that 4-1BB-AP was binding to proteins extrinsically attached to the cell surface (possible extracellular matrix components), B-cell lymphomas were washed in acid conditions prior to the binding assay. 4-1BB-AP still bound specifically to mature B-cell lymphomas. It is still to be determined whether a 4-1BB-ligand specifically expressed on B cells and macrophages exists, and whether 4-1BB-AP may bind to the ECM under particular binding conditions. It is possible that the ECM could facilitate the binding of 4-1BB to a specific cell-surface ligand.

B cells and helper T cells interact with each other through receptors on B cells binding to their specific counter-receptors on T cells. This interaction results in a cascade of biochemical signaling relays between the two cell types. As this interaction proceeds, these cells become committed to enter the S-phase of the cell cycle. Initial interactions between TCR and CD4+ on T cells, and processed antigen-MHC II on B cells, do not result in B cells capable of entering the cell cycle (Noelle and Snow et al., 1990). However, studies from in vitro systems suggest that once T cells are stimulated, they express newly synthesized or modified cell-surface molecules capable of inducing B cells to enter the cell cycle. This T cell function is not antigen-specific or MHC-restricted. In addition, soluble factors are not required for the Th induction of B-cell activation. Once B cells enter the cell cycle, IL-4 induces B cells to progress from G₁ to S phase. The ability of activated T cells or T-cell membranes to promote the entry of B cells into the cell cycle can be blocked by either cycloheximide or cyclosporin A treatment. These newly expressed membrane proteins appear to be "lymphokine-like" in their induction characteristics.

4-1BB has expression properties which meet the requirements of a B-cell co-stimulator. 4-1BB is inducible by anti-CD3 or TCR-mediated T-cell stimulation, and its expression is sensitive to cyclosporin A as well as cycloheximide treatment. Interestingly, paraformaldehyde-fixed SF21-4-1BB cells, synergized anti-.mu. and induced B-cell proliferation. The co-stimulation of splenic B cells by SF21-4A1BB occurred at optimal (10 .mu.g/ml) and sub-optimal (1.0-0.1 mg/ml) doses of anti-.mu.. The addition of SF21-4-1BB cells to resting B cells, did not result in significant B-cell proliferation. SF21-4-1BB cells did not synergize with tetradecanoylphorbol acetate (TPA) or ionomycin, or sub-optimal concentrations of LPS in inducing B-cell proliferation.

Although the baculovirus system has been used to express large amounts of recombinant soluble proteins, this system may be utilized for the expression of recombinant cell-surface proteins. The baculovirus infection provides a convenient means to express uniformity high levels of recombinant protein on a per cell basis. It is noteworthy, that the addition of SF21 cells alone did not result in significant levels of co-stimulation. This can be a potential problem when using COS- or L-cell lines which can exhibit strong co-stimulation activity on their own.

Another member of the NGFR superfamily, CD40, is expressed on B cells and interacts with gp39, a molecule expressed on activated T cells. The cDNAs encoding the murine and human gp39 proteins have been cloned; this cell surface molecule is a type II membrane protein with homology to tumor necrosis factor. Noelle et al., found that a CD40-immunoglobulin fusion protein, is capable of blocking T cell-induced B-cell proliferation and differentiation in a dose-dependent manner. Armitage et al. have isolated a cDNA for murine gp39 and showed that gp39 could induce B-cell proliferation in the absence of costimuli, and result in IgE production in the presence of IL4-. Hollenbaugh et al., have shown that COS cells transfected with human gp39 can synergize with either TPA or anti-CD20 in inducing human B-cell proliferation and is able to stimulate B cells without a co-stimulator only at low levels. These data indicate that CD40 may be one of the B-cell-surface molecules that transmit signals during physical contact with T cells.

Cell-surface receptors communicate with their external milieu by interacting either with soluble factors or other cell surface molecules expressed on neighboring cells. The role of biochemical signals delivered by cell-cell contact versus those delivered by soluble factors interacting with cell surface receptors is not clear. The NGFR superfamily is unusual for the TNFR I and II as well as the NGFR bind to more than one ligand. The TNFRs I and II both bind to TNF-a and TNF-R The NGFR binds to NGF, brain-derived neurotrophic factor, and neurotrophin-3.

In addition, one ligand may function as both a cell surface and soluble ligand. Recent evidence on the CD40 ligand, gp39, suggests that this ligand can exist as a membrane bound as well as a soluble ligand. It may be possible that 4-1BB is-secreted and interacts with B cells in a soluble form as well as a membrane bound form. A member of the NGFR receptor family, CD27, which is expressed on T cells, is secreted in addition to being expressed on the cell surface (Hintzen et al., 1991). It is also possible that more than one 1 ligand, if soluble and on the cell surface, may bind to 4-1BB.

Isolation of the Human Homologue, H4-1BB

In order to isolate the human homologue (H4-1BB) of mouse 4-1BB two sets of polymerase chain reaction (PCR) primers were designed. To design the PCR primers, the amino acid sequence among the members of nerve growth factor receptor (NGFR) superfamily were compared because 4-1BB is a member of the superfamily (Mallett and Barclay, 1991). The amino acid sequences employed were mouse 4-1BB, human NGFR, human tumor necrosis factor receptors, human CD40, and human CD27. The areas of sequence conservation among the NGFR superfamily were chosen.

Materials and Methods

Peripheral blood lymphocytes from normal healthy individuals were isolated and activated with PMA (10 ng/ml) and ionomycin (1 mM). Messenger RNA from the lymphocytes was isolated. Using reverse transcriptase the human lymphocyte mRNA was converted to single-stranded cDNA The cDNA was then amplified with Taq polymerase with combination of the primers (SEQ ID NOS. 3-6). A combination of primers was used and produced a specific band of .about.240 bp. The 240 bp is an expected size of human 4-1BB if the human homologue protein is similar to mouse 4-1BB in size. The PCR product (240 bp) was cloned in PGEM3 vector and sequenced. One open reading frame of the PCR product was 65% identical to mouse 4-1BB. Therefore, it was concluded that the 240-bp PCR product is the human homologue of mouse 4-1BB. The 240-bp PCR product was used to screen .lambda.gt11 cDNA library of activated human T lymphocytes., An 0.85 kb cDNA was isolated. The sequence of the cDNA is shown in (SEQ. ID NO:1) and the predicted amino acid sequence is shown in (SEQ. ID NO:2).

An expression plasmid to produce H4-1BB-AP fusion protein was constructed. The 5' portion of the H4-1BB cDNA including sequences encoding the signal sequence and the entire extracellular domain, was amplified by PCR. For correctly oriented cloning, a Hind III site on the 5' end of the forward primer and a Bg1 II site on the 5' end of the reverse primer were created.

The Hind III--Bg1 II H4-1BB fragment was inserted into the mammalian expression vector APtaq-1, upstream of the coding sequence for human placental alkaline phosphatase (AP).

H4-1BB-AP will be used to identify cells and tissues that express ligand for human 4-1BB (i.e. H4-1BBL). The studies with mouse 4-1BB indicated that the ligand for 4-1BB is on the cell surface. B cells and macrophages were major cells that express 4-1BBL. It is expected that H4-1BBL be expressed on human B cells and macrophages.

Results

A mammalian expression cDNA library was generated from human cell lines that express H4-1BBL. The library was screened by Iodine-labeled H4-1BB-AP. cDNA for H4-1BBL was then isolated and characterized. Soluble recombinant H4-1BBL was then produced. The generated antibodies were then used to suppress or enhance immune responses as described below. Monoclonal antibodies to H4-1BBL were produced and are discussed below.

According to studies completed by the inventor, 4-1BB acts as a co-stimulatory signal. It is expected then that H4-1BB acts as a co-stimulatory signal for T cell activation. Mouse 4-1BB helped B cells with proliferation and differentiation. H4-1BB has been found to do the same. H4-1BB-AP, H4-1BBL can be used to suppress or enhance human immune responses.

Claim 1 of 8 Claims

What is claimed is:

1. An isolated agonistic monoclonal antibody which binds to human 4-1BB (H4-1BB) having SEQ ID NO:2.

 

____________________________________________
If you want to learn more about this patent, please go directly to the U.S. Patent and Trademark Office Web site to access the full patent.