WO1994013312A1 - Mucosal vascular addressin, dna and expression - Google Patents

Abstract

Mammalian purified nucleic acids and proteins encoding mucosal addressins are provided. The proteins, nucleic acids and fragments thereof serve a variety of purposes in modulating leukocyte mucosal tissue homing interactions. In addition, the protein fragments may serve as antigens to produce antibodies specific for particular domains of the mucosal addressin, while the nucleic acids may be used as probes, for antisense preparation, for introduction into hosts which may be allogeneic or xenogeneic to the DNA source for expression of the protein, and the like.

Description

MUCOSAL VASCULAR ADDRESSIN, DNA AND EXPRESSION

INTRODUCTION Technical Field

The field of this invention is the encoding DNA and expression of mucosal vascular addressin.

Background

Blood is the highway of leukocytes, where the leukocytes can move from various tissues, particularly lymphoid tissues, and be transferred to other sites. Since many of the leukocytes monitor for foreign matter and diseased cells, it is necessary that they circulate in a continuous manner to provide protection from disease. Many leukocytes are also involved in providing factors for growth of other cells and aiding in regenerative processes. It is found, however, that certain subsets of leukocytes appear to be programmed to go to different sites. For example, cells will differentially home to mucosal tissue, peripheral lymph nodes, synovial tissue or skin, depending upon the particular proteins which are present on the cell surface. Thus, there are subsets of lymphocytes and monocytes which will preferentially be directed to particularly types of tissue.

The process of a cell being directed to a particular site is referred to as "homing. " What this means is that there is a process which allows the cell programmed for a particular site to be directed to that site in the event of injury or disease, or in some instances under normal physiologic circumstances. Part of this process involves endothelial cells, which are part of blood vessels, including high endothelial venules. These cells, present on venules, respond to signals from diseased or injured tissue by upregulating certain surface membrane proteins. The surface membrane proteins, referred to as addressins, are capable of specifically binding to particular surface membrane proteins of certain lymphocytes or monocytes, providing for the specificity of the circulating cells. After binding to the endothelial cells, the circulating cells may then be directed to the site of injury or disease by a process referred to as "extravasation."

There is substantial interest in being able to control the homing process. There are instances where one would wish to inhibit transfer of cells to a particular site, such as in the case of autoimmune diseases. Alternatively, there may be substantial interest in directing leukocytes to a particular site or for them to be maintained at a particular site, where it may be desirable to concentrate a population of T-lymphocytes or neutrophils. Also, there may be an interest in directing particular compositions to cells or sites for therapeutic or diagnostic purposes. There is, therefor, substantial interest in identifying proteins associated with the homing process, where these proteins may then aid in inhibiting or directing leukocytes or compositions to a designated site in the body.

Relevant Literature

The mechanism of homing of lymphocytes is described by Butcher et al. , Eur. J. Immunol. 10, 551-556; Butcher, Curr. Topics Microbiol. Immunol. 133, 2961-2965 (1986); Picker and Butcher, Annual Rev. Immunol. 10, 561-591 (1992); and Springer, Nature 346, 425-434 (1990). See also, Streeter et al., Nature 331, 41-46 (1988) and Nakachli et al., Nature 337, 179-181 (1989). The murine brain endothelioma bEnd3 is described by Montesano et al., Cell 62, 435-445 (1990). Descriptions of homing receptor and addressin interactions may be found in Berg et al., Academic Press_. San Diego, 111-129 (1991).

SUMMARY OF THE INVENTION Nucleotide sequences are provided encoding mammalian mucosal addressin, where the sequences can be used for expressing the addressin or fragments thereof. The addressin or fragments thereof may be used to produce antibodies which may find therapeutic purposes. The addressin or fragments thereof, by themselves or joined to other molecules, may find use in tagging leukocytes, inhibiting leukocyte binding to endothelial cells, in isolating leukocytes, in transporting various agents to leukocytes, and the like. Also as a tool for screening to identify molecules that bind to the mucosal addressins, that might themselves find therapeutic uses in inhibiting MAd functions, or might be used to target other agents to MAdCAM-1 and endothelial cells ("EC") for therapeutic purposes. Methods are provided for isolating or synthesizing DNA which encodes at least a portion of the mucosal addressin, for introducing the DNA into host cells, and for expression of the addressin or fragments thereof. In addition, the addressin can serve as a source of the carbohydrate side chains which serve as binding entities to the leukocytes.

BRIEF DESCRIPTION OF THE FIGURES

The Figure depicts the cDNA sequence of murine MAdCAM-1 (Clone 7) and the encoded amino acid sequence. The signal peptide, transmembrane region and single amino acid change (R to L) are bold underlined. In the nucleic acid sequence the base change is T to G (nucleotide 182).

DESCRIPTION OF THE SPECIFIC EMBODIMENTS Purified nucleic acid and protein compositions are provided, where the protein compositions are mammalian mucosal addressins or fragments thereof and the nucleic acids are sequences encoding the mucosal addressin or fragments thereof. The mucosal addressin protein in the mouse is characterized by binding to antibodies which have been shown to bind to mucosal addressins, MECA-89 and -367. In addition, the mucosal addressin has a structure having three regions, the extracellular region, the transmembrane region, and the cytoplasmic region and when initially expressed, the signal sequence. The mature protein has from about 350-425 amino acids, usually from about 375-400 amino acids. At least about 80% of the mature protein will be the extracytoplasmic portion.

The overall structure of the protein is a Type I transmembrane protein. There are three immunoglobulin-related domains. The most end-terminal domain (domain 1) shows high sequence similarity to domain 1 of the cell adhesion molecule ICAM-1. The second immunoglobulin domain (domain 2) shows sequence similarity to the fifth domain of mouse VCAM-1. Domains 1 and 2 are separated by at least 20 amino acids and not more than about 50 amino acids, usually about 35-45 amino acids. This is a typical spacing for separation of two adjacent immunoglobulin-like domains.

Just proximal to domain 2 is a long stretch rich in prolines, serines and threonines, indicating a site for heavy O-glycosylation. This domain will be at least about 60 amino acids and not more than about 100 amino acids, generally being from about 65-80 amino acids. Of this domain, the majority of serines and threonines are found in about a 35-40 amino acid stretch. In this region, from about 30-50% of the amino acids will be serine or threonine, while from about 5-20% of the amino acids will be proline. Proximal to this region is a third immunoglobulin domain, domain 3. This domain has significant homology with a constant region immunoglobulin domain C2 of the IgAl heavy chain. It should be noted that IgAl is associated with mucosal tissue.

The mucosal addressin may come from any species, particularly primate, more particularly human, rodentiae, particularly murine, namely mouse and rat, lagomorpha, ovine, bovine, equine, porcine, feline, canine, guinea pig, etc.

The mouse mucosal addressin is characterized by having a signal sequence of 21 amino acids, an extracytoplasmic region of 342 amino acids, a transmembrane sequence of 22 amino acids and a cytoplasmic tail of 20 amino acids. The extracellular region has three potential N-linked glycosylation sites, which do not appear to be used.

The mucosal addressins have substantial homology to other adhesion molecules which have been previously reported, namely VCAM-1 and ICAM-1. This homology is particularly found at the N-terminal domain. The subject mucosal addressin will hereinafter be referred to as

MAdCAM-1.

The subject protein is selectively expressed on HEV of mucosal lymphoid organ HEV and lamina propria venules. The subject protein binds to the previously-reported monoclonal antibodies MECA-89 and -367 and specifically binds to the mucosa-selective T-cell lymphoma TKl. MAdCAM-1 is a member of the immunoglobulin superfamily.

The gene encoding MAdCAM-1 will generally be joined to other than wild- type sequences. Usually the wild-type DNA will be not more than about 1 cM, generally not more than about 10 kb. The genomic gene will usually be separated from the intact chromosome and may be manipulated as part of a YAC, where fragments will be from about 100-1000 kbp.

The cDNA which encodes the subject protein in the mouse has between 1400-1500 nucleotides (nt) with a short, 5'-noncoding region, generally from about 20-40 nt and a 3'-noncoding region of from about 150-250 nt. The mouse cDNA is 1436 nt, with a 5'-noncoding region of 30 nt, and a 3'-noncoding region of 191 nt. Included in the 3' region is a polyadenylation site at position 1400. The initiation codon, ATG, is at position 31 of the cDNA. The molecule encoded has 21 amino acids for the signal peptide leaving a mature polypeptide of 384 amino acids.

By manipulation of the coding sequence, variations of the mucosal addressin may be achieved. Of particular interest is removal of the transmembrane sequence, either with or without removing the cytoplasmic portion, whereby a soluble secreted form of the mucosal addressin will be obtained. Excising the transmembrane region may be achieved by in vitro mutagenesis, polymerase chain reaction, or other known technique.

Expression of the mucosal addressin in endothelial cells, particularly transformed endothelial cells, can be enhanced by using TNF- or lipopolysaccharide, with relatively high levels of expression occurring within a few hours, generally within about 4-6 hr.

Expression of the protein in an appropriate mammalian host will provide for the glycosylated mature product, which may be isolated and purified in accordance with conventional ways. The protein can be obtained in a pure state of at least about 50 wt. % , preferably at least about 75 % and more preferably substantially pure or pure, in that the composition is free of any physiologically active contaminants.

The various compositions provided by the subject invention may be used in a variety of ways. The intact protein, fragments of the protein indicating polypeptide fragments, polysaccharides, and combinations thereof may find application. The DNA encoding the mucosal addressin, the entire reading frame, with or without introns, non-coding regulation elements of the gene, or fragments of the encoding DNA, particularly fragments encoding all or a portion of a functional region or domain can find use. The sequences may be mutated for a variety of reasons, where the mutations may include insertions, deletions and substitutions. Substitutions may be conservative or non-conservative, there will usually be fewer than the greater of 2 substitutions or substitutions equal to 2% of the number of nucleotides. For the protein, the same numbers will apply for the amino acids.

Polypeptides of each of the domains may be used to prepare a specific antisera or monoclonal antibodies for a specific site. In this way, binding to each of the domains may be individually inhibited by using antibodies which specifically bind to that domain. These antibodies may then be used to determine the manner in which binding occurs between the addressin and various homing receptors or selectins. The glycosylated domain may be used in a variety of ways, such as providing a source of the sugars which may bind to lectin regions of the selectins, for producing antibodies to the protein and/or saccharide portions of the domain, to serve as inhibitors to binding, to direct compositions to activated leukocytes, and the like.

The protein or fragments thereof, usually fragments of at least about 12 amino acids, more usually at least about 16 amino acids and preferably at least about 20 amino acids, generally fewer than the entire protein, preferably fewer than 60 amino acids may be employed as agents to block binding of the leukocytes to the mucosal addressin. These peptides may find use in vitro and in vivo, where the peptides may inhibit entry of leukocytes to a site where their presence may contribute to a disease. Thus, one may wish to exclude leukocytes from injured sites, where the leukocytes may enhance inflammation, induce cellular proliferation, destroy native tissue, or the like. Diseases which may be treated include especially intestinal inflammatory disorders including Crohn's disease, ulcerative colitis, regional enteritis, and Celiac disease; but also in inflammatory and autoimmune disorder in other sites where the MAdCAM-1 may be induced abnormally or in pathologic conditions. Also in malignancies, especially of intestinal carcinomas and their metastases, in which MAdCAM-1 may be induced in tumor-associated vessels.

The protein or fragments thereof may be modified in a variety of ways for a variety of purposes. The peptides may be joined to immunogens to enhance immunogenicity for the formation of antibodies. The antibodies may be from any convenient source but will normally be produced by laboratory animals, such as mice, rats and rabbits, or in domestic animals, such as cows, sheep, pigs, etc. The manner of producing antisera and monoclonal antibodies is sufficiently well- established as to not require description here. The mouse or other species monoclonal antibodies may be humanized by replacing the constant region, by itself or in combination with one or more conserved framework regions, with regions from human antibodies, so that human IgA, -G, or -M or the like regions may replace the native antibody regions to reduce antigenicity in humans and to provide effector functions.

The protein or peptides may be joined to polyethylene glycol, constant regions of immunoglobulins, e.g. IgG, to lipids or the like, to enhance the lifetime of the peptide or protein when administered in vivo. The peptide or protein may be joined to various effector agents, particularly agents which produce a physiological reaction, such as antibodies, toxins, drugs, carbohydrates or cytokines. The peptide or protein may be joined to these various agents in a variety of ways. Where proteins are involved, the sequence encoding the subject peptide or protein may be fused to the other protein to provide for a fusion protein. Where a lipid is to be joined, it may be feasible to provide for the coding sequence which encodes for the linking of the peptide to a lipid through a phosphate linkage, based on known, natural processes of the host cell. Alternatively, various chemical linking agents are known which allow for linking through carboxyl groups, amino groups, thiol groups, or hydroxyl groups, as well as linking through an activated aromatic group, such as is present in tyrosine and tryptophan. A description of these agents may be found in Antibodies, A Laboratory Manual, Harlow, E., Lane, D. ed., Cold Spring Harbor Laboratory, 1988, pp. 319-358. Various synthetic procedures and conditions are known, which need not be described here.

In addition, the peptides or proteins may be labeled with a wide variety of labels for a variety of purposes. Thus, the peptides or proteins may be used in assays for detecting the presence of cells binding to the subject mucosal addressin or fragment thereof. Thus, fluorescent labels, enzyme labels, radioactive labels, or the like may be employed for use of the subject peptides or proteins in the assays. Numerous assays are known, such as RIA, ELISA, fluorescence polarization assay, fluorescence-activated cell sorters (FACS), and the like.

In addition, the peptides or proteins may be bound to a solid support, such as particles, microtiter plate wells, solid surfaces, where the peptides may serve to bind to cells having the appropriate complementary protein. In this manner, cells may be selectively removed ex vivo for isolation and characterization, to reduce their population in a host, or the like. The particles may be magnetic particles, pore glass beads, latex particles, agarose particles, sepharose particles, or the like. The various particles may be activated in a wide variety of ways, activated particles being commercially available for covalently linking the peptide or protein to the particle. Similarly, other surfaces may also be activated by known ways for covalent binding. Alternatively, depending upon the purpose of the binding, with some surfaces, e.g. polystyrene, non-covalent binding may suffice. After binding of the subject peptide or protein, usually an innocuous protein will be used to react with any remaining unreacted functional groups and to coat any hot spots which might result in non-specific binding in a subsequent application.

The DNA encoding the subject addressin or fragment thereof may be used for a diversity of purposes. The cDNA or fragment thereof may be used to identify the genomic gene which encodes the subject addressin. A cDNA or fragment thereof from one species may be used to probe cDNA or genomic libraries of other species to identify and isolate the gene from the other species. In addition, one may use the sequence encoding the mucosal addressin to define antisense sequences, which may then be used to inhibit expression of the gene. In this way, one can inhibit the expression of the mucosal addressin in mucosal endothelial cells, thus inhibiting the homing of lymphocytes via these endothelial cells.

One may also use the sequences for homologous recombination, so that the sequences may be used to correct a defective mucosal addressin gene, to specifically knock-out a mucosal addressin gene or to provide for enhanced or regulated expression of the mucosal addressin gene, by changing the promoter or other regulatory sequence present upstream from the coding region. In this way, one can investigate the effect on endothelial cells of the inability to express the mucosal addressin. The subject peptides may be used in assays to identify drugs, including peptides, saccharides and small synthetic organic molecules which bind to the peptides and have biological activity. These agents may then be used in assays to determine their effectiveness in vivo. The initial assays may take various forms involving direct binding or competitive binding. In the former, the subject peptides may be bound to a surface and the radiolabeled agent added. By determining the radioactivity bound to the surface after washing to remove non¬ specific binding, the affinity of the agent may be determined. Alternatively, one can add the agent to a complex of a known binding agent and the peptide and determine the off rate or equilibrium concentration of the labeled known binding agent as an indication of the affinity of the candidate agent.

The proteins or fragments thereof, capable of binding to the complementary binding protein may also be used as antagonists for complex formation and to modulate the various interactions. Thus, by administering the respective protein or fragment thereof to a host, the protein may serve to home to the complementary binding member and inhibit the binding of the addressin associated with the target cells. The fragment will usually be at least 8 amino acids, more usually at least about 10 amino acids, frequently at least about 16 amino acids, usually not exceeding about 60 amino acids, preferably not exceeding about 36 amino acids. Rather than acting as inhibitors or enhancers to prevent or increase complex formation between leukocytes and target sites, the proteins, fragments thereof, or anti-idiotope antibodies may serve to direct a wide variety of molecules to the target site. Thus, in the case of neoplastic tissue, by administering one of the subject compounds or compositions bound to a therapeutic drug, one can direct the binding of a therapeutic drug to the desired site for retention, and concentration at the desired site to which the MAdCAM-1 expressing endothelial cells bind. One can provide for the binding of radioisotopes for in vivo diagnosis or imaging, for radiotherapy, or the like. Alternatively, one could bind cytotoxic drugs to the fragments to kill target lymphocytes. In administering the various therapeutic agents, for the most part, empirical determinations will be involved to determine the level of therapeutic agent. The level of therapeutic agent which is administered will depend to a substantial degree on the stability of the protein, its size, the manner of administration, the site of administration, the purpose of the therapy, and the like. Therefore, no simple range may be given which would indicate what levels should be applied for any particular therapy. For the most part, the proteins will be administered in an appropriate physiologically-acceptable medium, e.g. water, saline, phosphate- buffered saline, mineral oil, Ringer's solution, or the like. Administration will normally be parenteral, particular intravascularly, but may be oral. For the reasons given above, the course of treatment will also vary.

There may be situations where there is an interest in using cells as therapeutics. By employing the subject peptides, one may isolate particular subsets of cells, which may then be expanded in culture in an appropriate nutrient medium or in co-culture, and may then be used for a therapeutic purpose. Depending upon the particular situation, the cells may be autologous or allogeneic.

The antibodies may also find wide use for detecting the presence of venules in vitro or in vivo, where the subject mucosal addressin is upregulated. By using antibodies, binding of leukocytes to these sites may be inhibited, so as to prevent extravasation to the particular site. The antibodies may also be used to make anti-id antibodies, which may be selected as mimics for the subject peptides or proteins. In this way, the antibodies which will have a substantially different composition from the mucosal addressin, may fulfill the same function by binding to the endothelial cell and block binding of the leukocyte to the venule. Also, the antibodies may be used in the variety of ways indicated above in place of the subject peptides or proteins.

The cDNA or genomic DNA encoding MAdCAM-1 can be used in preparing expression constructs. The DNA can be used in any convenient host, both prokaryotic and eukaryotic for the cDNA and invertebrate and vertebrate for splicing capability for the genomic gene. As previously indicated, for glycosylation, eukaryotic, particularly mammalian hosts will be required. The gene encoding MAdCAM-1 may be inserted into a wide variety of generally- available expression cassettes having linkers between a 5' transcriptional and translational initiation region, which may include an enhancing sequence, and may provide for constitutive or inducible transcription, and a 3' transcriptional and translational termination region, which will normally include a polyadenylation signal. Alternatively, one may prepare an expression cassette by combining the appropriate functional sequences in the appropriate order as indicated above. Numerous transcriptional initiation regions are commercially available or described in the literature. The expression cassette may then be introduced into a vector for replication and expansion. The resulting vector will usually include a marker for selection of hosts containing the construct. The vector may be introduced into the expression host by any convenient means, precipitated DNA, transfection, fusion, electroporation, etc., where the vector may provide for extrachromosomal maintenance or integration into the genome of the host. Of particular interest are vectors comprising sequences of a virus for transfection and integration into mammalian cellular hosts. The expression host may then be grown. Depending on the host, the product may be secreted and isolated from the supernatant or in the event it is retained in the cytoplasm, the cells may be harvested and the lysate extracted. The product may be purified using electrophoresis, gel chromatography, affinity chromatography, extraction and the like. The following examples are offered by way illustration and not by way limitation.

EXPERIMENTAL A transformed mouse endothelioma cell line, Bend3 cells was employed. It was shown to be inducible for the MECA-89 and -367 epitopes with bacterial lipopolysaccharide (LPS) and tumor necrosis factor (TNF-α). Serial fluorescence activated cell sorting was used to generate a variant line termed Hi-MAd 4 that expresses much higher levels of the antigen upon TNF-α stimulation. Hi-MAd 4 cells, stimulated with recombinant TNF-α for either 6 or 8 hours were pooled and used as the source of RNA for cDNA libraries. Endothelial cell cDNA libraries were synthesized with a BRL system which generates directionally-oriented cDNA into the λgtll derivative, λgt22A. Phage clones (5 X 10⁵) were screened with a mixture of anti-mucosal addressin antibodies consisting of a rat polyclonal against affinity-purified MAd (1:2000) and the monoclonal antibodies MECA-89 and -367 (5 g/ml) diluted TBS with 1 % BSA and 0.02% sodium azide. Positive clones were detected with an anti-rat IgG conjugated to alkaline phosphatase (Promega) and BioRad BCIP/NBT color detection reagents. The entire sequence of the mucosal addressin cDNA is 1436 nt in length, consisting of a 30 nt 5' untranslated region and a 191 nt 3' untranslated region. A canonical polyadenylation signal AATAAA is located at position 1400. The longest open reading frame of 405 amino acids begins with the first ATG at position 31 and ends with the termination codon TGA at position 1245. Hydrophobicity analysis and rules governing signal peptide cleavage patterns predict a signal peptide of 21 amino acids, leaving a mature polypeptide of 384 amino acids. This sequence predicts a typical transmembrane protein consisting of a 342 amino acid extracellular domain, a 22 amino acid transmembrane region and a short, charged 20 amino acid cytoplasmic tail.

The extracellular domain contains 3 potential N-linked glycosylation sites with the consensus sequence NXT/S at position 718, 787 and 111. Studies indicate that these sites are not utilized, since the mucosal addressin is insensitive to peptide N-glycosidase. The largest cDNA inserts from λ clones 7 and 15 (λ MAd-7 and λ MAd-15) were subcloned into Bluescript vectors (Stratagene) to construct the plasmids pMAd-7 and pMAd-15. Plasmids for sequencing were initially generated by making nested deletions with exonuclease III using Erase A Base (Promega) system. The DNA sequence was determined by the dideoxy method using fluorescent-labeled universal primers and the Applied Biosystems DNA sequence analyzer. Gaps in the sequence were filled in by subcloning small restriction fragments into appropriately-digested Bluescript vectors. Two independent clones were sequenced 100% on both strands and are identical with the exception of a single base substitution which would generate a leucine instead of the arginine shown in pMAd-15 at amino acid 61.

In the sequence provided in the Figure, the putative hydrophobic signal peptide and transmembrane sequences have a bold underline. Potential N-linked glycosylation sites are present at positions 718, 787 and 111. Cysteines likely to form disulfide bonds in Ig domains are boxed while other cysteines are underlined. The 37 amino acid "mucin domain" between amino acid residues 221 and 257 is outlined with a bold line. The polyadenylation signal ATAAA at position 1400 is indicated in bold. Poly A+ RNA from unstimulated high MAd4 cells, high MAd-4 cells stimulated with TNF-α for 6-8 hrs, high MAd-4 cells stimulated with TNF-α for 20 hrs, the Ll-2 pre-B lymphoma, normal 8 wk. BALB/c mouse liver, kidney, Peyer's patches, brain, peripheral lymph nodes, lung, mesenteric lymph nodes and spleen were hybridized to the 1.4 kb MAd cDNA insert. Northern blots used 1 μg of poly A+ RNA from cells lines and 2 μg from tissues which was denatured and electrophoresed through a 1 % agarose formaldehyde gel and transferred to a PVDF (Immobilon, Millipore) membrane by standard capillary blot procedures. Equal amounts (by OD₂₆₀) of each RNA were electrophoresed and stained with ethidium bromide to ensure equivalent loading of each sample. Hybridization was performed at 65°C using standard conditions. cDNAs were labeled with α³²P dCTP by priming with random hexamers.

Northern blot analysis showed that two species of RNA at approximately 1.8 and 1.3 kb were slightly induced in high MAd-4 cells stimulated with TNF-α for 6-8 hrs and were highly induced by 20 hrs. The same transcripts are highly expressed in mesenteric lymph nodes (MLN) and Peyer's patches (PP) and detected at lower levels in spleen and peripheral lymph nodes (PLN). The species were not detected with the pre-B lymphoma Ll-2 or in liver, brain, kidney or thymus. This pattern of expression is consistent with previous immunohistological studies which have shown high expression of MAd in HEV, in PP, mesenteric lymph node (MLN) and some expression in the marginal sinus around splenic white pulp nodules in the spleen. A distinct transcript of approximately 1.5 kb was detected in unstimulated high MAd-4 cells, Ll-2 cells, kidney and liver, implying either a related gene or alternatively spliced product in these tissues. The expression construct pCDM15 was generated by subcloning the

Sall-Notl insert of λMAd-15 into XhoI-NotI digested pCDM8. This construct was transfected onto COS cell monolayers (along with no DNA as a mock transfected control) using DEAE dextran. After 48 hrs., cells were replated into 3.5 cm dishes to ensure equal plating of all transfectants and 24 hrs. later transfectants were analyzed by immunofiuorescence. Both addressin and mock transfectants were stained with anti-addressin monoclonals MECA-89 or -367 alone or the isotype matched control MEL-14. Staining was detected both visually and by FACS analysis using a phycoerythrin-conjugated goat F(ab')₂ anti rat IgG (Caltag). Transfection efficiencies were approximately 25% for each experiment.

For adhesion assays, COS cell transfectants were plated into 3.5 cm plates as described above. B and T lymphomas were pelleted and resuspended at a density of 2 X 10⁶ per ml in RPMI without sodium carbonate, with 2% serum and buffered with 10 mM HEPES, pH 6.7. Cell suspensions were layered (1 ml per plate) onto addressin and mock transfectants and gently rotated on a gyrotary shaker (New Brunswick Scientific Model G-2) at 60 rpm for 30 min. at 25 °C, and then washed 4 times in the same medium by aspirating and gently filling each plate with 1 ml per wash. Assays were fixed in Dulbecco's PBS containing 1 % glutaraldehyde and sodium azide. For antibody inhibition assays, transfectants were pre-treated with 20 μg/ml of anti-addressin antibody MECA-367 or isotype matched control 9B5 (anti-CD44) at 1 ml per plate for 20 min. at 25 °C. Solutions were then aspirated and adhesion assays were performed as described above. The average number of cells bound for each experiment was determined by counting cells in 4 random fields.

Both immunofiuorescence and FACS analysis showed that COS cells transfected with pCDM15 (the full-length addressin cDNA inserted into the CDM8 expression vector) expressed antigen which reacts with anti-mucosal addressin MAbs MECA-89 and -367, but not with isotype matched control MAbs. To assay the function of the cloned MAd cDNA, adhesion of the mucosal HEV binding lymphoma cell line TKl to the COS cell transfectants showed that TKl cells bind to pCDM15 transfected COS cells, but not to control (mock) transfectants. Binding is specific to cells expressing MAd as shown by parallel immunofiuorescence with a non-blocking MAd, MECA-89. Adhesion is inhibited by approximately 95 % by pre-treatment with saturating levels of anti-addressin MAd MECA-367, while pre-treatment with control MAd had no effect. Furthermore, binding was specific for a mucosal HEV binding cell line, as Ll-2 and NS8 cells (lines previously shown not to adhere to HEV) failed to adhere to the COS cell transfectants.

The amino acid sequence of the mucosal addressin was analyzed for homology to other proteins in the NBRF database using the program FASTP and the highest degree of homology was to murine ICAM-1. Consistent with the designation of ICAM-1 as a member of the immunoglobulin supergene family, the extracellular domain of MAd can also be folded into a structure containing 3 immunoglobulin-like domains. Each of the 3 domains contains an invariant pair of cysteine residues known to stabilize the Ig loop, with distances of 44, 66 and 47 amino acids between cysteines, respectively.

Additional searches with the individual immunoglobulin-like domains of MAd revealed the highest homology of the most N-terminal domain (domain 1) with the first domain of rat ICAM-1 and the first domain of human VCAM-1. The first domain in all three molecules is similar in that the spacing between the cysteine residues and the predicted β-strand structure is typical of H or C2 immunoglobulin domains. Additionally, all three molecules have the unusual feature of double cysteine residues at both ends of the domain. The N-terminal Ig domains of ICAM-1 and VCAM-1 have been shown to contain binding sites for their respective lymphocyte counter-receptors LFA-l/a_rjSj and VLA-4/α₄jδ₁. The α₄jδ₇ integrin is known to be involved in lymphocyte binding to Peyer's patch-HEV in in vitro assays. The present finding of strong homology to other immunoglobulin-related vascular integrin receptors, supports an integrin binding motif lying in this domain in MAdCAM-1.

The second immunoglobulin domain is most closely related to the fifth domain of mouse VCAM-1. The third immunoglobulin-like domain of the MAd is unique in that the most significant homology is to the CH2 region of human and gorilla IgAl (33% identity). This homology relates two molecules which are highly expressed in mucosal tissues and which both interact with T-cells involved in mucosal immunity. A discrete subset of T-cells, enriched in Peyer's patches, constitutively expressed Fc receptors capable of binding IgA. Their binding to mucosal addressin could be strengthened by interaction of the IgA F_c receptor or other T-cell surface receptors, which are capable of binding the IgA-1 homologous domain.

A proline/serine-threonine-rich region in the amino-terminal portion of this domain which aligns with the hinge region encoded with the IgAl CH2 exon may add flexibility to the addressin and its interaction with ligands on the surface of lymphocytes. The 37 amino acid region that lies between domains 2 and 3, which is extremely rich in its serine/threonine content (41 %), exhibits the sequence characteristic of mucins. This structure is characteristic of O-glycosylation sites. These domains are suggested to form rigid rods without secondary structure and may serve to extend ligand binding domains further above the cell surface, thus allowing more efficient interactions. Rosen and colleagues (Lasky et al., Cell 69, 927-938 (1992)) have described a 50 kb mucin-like protein GlyCAM-1, which is highly expressed in peripheral and mesenteric lymph nodes. The protein core of GlyCAM-1 is also serine/threonine rich and is proposed to serve as a backbone molecule capable of presenting a carbohydrate ligand(s) characterizing the peripheral lymph node addressin. These carbohydrate structures are recognized by the L-selectin lymph node homing receptor and by anti-PLN addressin MAb MECA-79. As MECA-79-defined epitopes associated with the peripheral lymph node addressin also decorate a portion of MAdCAM-1 expressed in mesenteric lymph nodes, this mucin-like domain in the mucosal addressin can serve as a site for addition of selectin-binding carbohydrates important for recruitment of lymphocyte subsets to the mesenteric lymph node and also, at a lower level, to the Peyer's patches. Proteoglycans can also bind and present heparin-binding cytokines at the cell surface. Thus, when appropriately modified, the mucin-like region may also be available to bind adhesion-promoting cytokine-activating factors required for triggering activation-dependent lymphocyte sticking to the HEV.

It is evident from the above results, that the subject invention provides an opportunity for modulating homing to mucosal tissue by a great diversity of compositions and protocols. In addition, the compositions can be used for diagnostic and therapeutic purposes, for isolation of subsets of lymphocytes or other leukocytes, and the like.

All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: THE BOARD OF TRUSTEES OF LELAND STANFORD

JUNIOR UNIVERSITY

(ii) TITLE OF INVENTION: MUCOSAL VASCULAR ADDRESSIN, DNA AND EXPRESSION

(iii) NUMBER OF SEQUENCES: 2

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Bertram I . Rowland

(B) STREET: 4 Embarcadero Center

(C) CITY: San Francisco

(D) STATE: California

(E) COUNTRY: USA

(F) ZIP: 94111

(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: PCT/US93/

(B) FILING DATE: 23-NOV-1993

(C) CLASSIFICATION: 530

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Rowland, Bertram I.

(B) REGISTRATION NUMBER: 20,015

(C) REFERENCE/DOCKET NUMBER: FP-57452/BIR;STAN-144

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (415) 781-1989

(B) TELEFAX: (415) 398-3249

(2) INFORMATION FOR SEQ ID NO:1 :

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1434 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 31..1245 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

CCACGCGTCC GTCACAAGAC AGAGGCAGGC ATG GAA TCC ATC CTG GCC CTC CTG 54

Met Glu Ser lie Leu Ala Leu Leu 1 5

CTG GCC CTG GCC CTA GTA CCC TAC CAG CTC AGC AGA GGA CAG TCC TTC 102 Leu Ala Leu Ala Leu Val Pro Tyr Gin Leu Ser Arg Gly Gin Ser P e 10 15 20

CAG GTG AAC CCC CCT GAG TCT GAG GTA GCT GTG GCC ATG GGC ACA TCC 150 Gin Val Asn Pro Pro Glu Ser Glu Val Ala Val Ala Met Gly Thr Ser 25 30 35 40

CTC CAG ATC ACC TGC AGC ATG TCC TGT GAC GAG GGT GTA GCC CGG GTG 198 Leu Gin lie Thr Cys Ser Met Ser Cys Asp Glu Gly Val Ala Arg Val

45 50 55

CAC TGG CGT GGT CTG GAC ACC AGC TTG GGC AGT GTA CAG ACC CTC CCA 246 His Trp Arg Gly Leu Asp Thr Ser Leu Gly Ser Val Gin Thr Leu Pro 60 65 70

GGC AGC AGT ATC CTC TCT GTA CGG GGC ATG CTG TCA GAC ACA GGC ACT 294 Gly Ser Ser lie Leu Ser Val Arg Gly Met Leu Ser Asp Thr Gly Thr 75 80 85

CCT GTG TGT GTG GGC TCC TGC GGG AGT CGA AGC TTC CAG CAC TCC GTG 342 Pro Val Cys Val Gly Ser Cys Gly Ser Arg Ser Phe Gin His Ser Val 90 95 100

AAG ATC CTT GTG TAT GCC TTC CCA GAC CAG CTG GTG GTG TCC CCG GAG 390 Lys lie Leu Val Tyr Ala Phe Pro Asp Gin Leu Val Val Ser Pro Glu 105 110 115 120

TTC CTT GTA CCT GGA CAG GAC CAG GTG GTG TCC TGC ACG GCC CAC AAC 438 Phe Leu Val Pro Gly Gin Asp Gin Val Val Ser Cys Thr Ala His Asn

125 130 135

ATC TGG CCT GCA GAC CCG AAC AGT CTC TCC TTT GCC CTG CTA CTG GGA 486 lie Trp Pro Ala Asp Pro Asn Ser Leu Ser Phe Ala Leu Leu Leu Gly 140 145 150

GAG CAG AGA CTG GAG GGT GCC CAA GCC CTG GAA CCA GAG CAA GAA GAG 534 Glu Gin Arg Leu Glu Gly Ala Gin Ala Leu Glu Pro Glu Gin Glu Glu 155 160 165

GAG ATA CAA GAG GCT GAG GGC ACA CCA CTG TTC CGA ATG ACA CAA CGC 582 Glu lie Gin Glu Ala Glu Gly Thr Pro Leu Phe Arg Met Thr Gin Arg 170 175 180

TGG CGG TTA CCC TCC CTG GGG ACC CCT GCC CCT CCT GCC CTT CAC TGC 630 Trp Arg Leu Pro Ser Leu Gly Thr Pro Ala Pro Pro Ala Leu His Cys 185 190 195 200

CAG GTC ACC ATG CAG CTG CCC AAA CTG GTG CTG ACC CAT AGA AAG GAG 678 Gin Val Thr Met Gin Leu Pro Lys Leu Val Leu Thr His Arg Lys Glu

205 210 215 ATT CCA GTA CTA CAG AGC CAG ACC TCA CCT AAG CCC CCC AAC ACG ACC 72 lie Pro Val Leu Gin Ser Gin Thr Ser Pro Lys Pro Pro Asn Thr Thr 220 225 230

TCT GCT GAG CCC TAC ATC CTG ACC TCA TCA AGT ACT GCT GAG GCA GTC 77 Ser Ala Glu Pro Tyr lie Leu Thr Ser Ser Ser Thr Ala Glu Ala Val 235 240 245

TCC ACT GGG CTC AAC ATC ACC ACC CTA CCT TCT GCC CCT CCA TAC CCC 82 Ser Thr Gly Leu Asn lie Thr Thr Leu Pro Ser Ala Pro Pro Tyr Pro 250 255 260

AAG CTT AGC CCT AGG ACT CTG AGC TCT GAG GGA CCT TGC CGC CCG AAA 87 Lys Leu Ser Pro Arg Thr Leu Ser Ser Glu Gly Pro Cys Arg Pro Lys 265 270 275 280

ATC CAC CAG GAC CTG GAG GCA GGC TGG GAG CTA CTC TGT GAA GCA TCC 91 lie His Gin Asp Leu Glu Ala Gly Trp Glu Leu Leu Cys Glu Ala Ser

285 290 295

TGT GGG CCC GGA GTT ACT GTG CGC TGG ACC TTG GCT CCT GGC GAC CTG 96 Cys Gly Pro Gly Val Thr Val Arg Trp Thr Leu Ala Pro Gly Asp Leu 300 305 310

\. GCA ACC TAC CAC AAG AGG GAG GCT GGG GCC CAG GCA'- GG CTA AGC GTG 101 Ala Thr Tyr His Lys Arg Glu Ala Gly Ala Gin Ala Trp Leu Ser Val 315 320 325

CTG CCC CCA GGT CCC ATG GTA GAG GGC TGG TTC CAG TGC CGC CAG GAC 106 Leu Pro Pro Gly Pro Met Val Glu Gly Trp Phe Gin Cys Arg Gin Asp 330 335 340

CCT GGC GGG CAG GTG ACC AAT CTG TAT GTT CCT GGC CAG GTG ACC CCG 111 Pro Gly Gly Gin Val Thr Asn Leu Tyr Val Pro Gly Gin Val Thr Pro 345 350 355 360

AAT TCC TCC TCC ACC GTC GTC CTA TGG ATT GGC AGC TTG GTG CTG GGG 115 Asn Ser Ser Ser Thr Val Val Leu Trp lie Gly Ser Leu Val Leu Gly

365 370 375

CTG CTT GCA CTG GTC TTC CTT GCC TAC CGC CTG TGG AAA TGC TAC CGG 120 Leu Leu Ala Leu Val Phe Leu Ala Tyr Arg Leu Trp Lys Cys Tyr Arg 380 385 390

CCA GGT CCT CGC CCA GAC ACT AGC TCA TGT ACA CAC CTA TGAAGCTCCA 125 Pro Gly Pro Arg Pro Asp Thr Ser Ser Cys Thr His Leu 395 400 405

TTATGCCAGA CTAAAGGAGG CAGAGAGTGA CCAGCTGCAG GATTTGGGGC ATCAAGATGA 131

TAGTGTGGCC TCTTTCCTTG GTGGTCAGCA CATCTATAAG TTTCTCCTGA CTTCTGGGCT 137

TTTCTGCCTG CTGGCCCAGA GCTAAATAAA AGCCCCGTAT CTAAAAAAAA AAAAAAAAA 1434 (2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 405 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Met Glu Ser lie Leu Ala Leu Leu Leu Ala Leu Ala Leu Val Pro Tyr 1 5 10 15

Gin Leu Ser Arg Gly Gin Ser Phe Gin Val Asn Pro Pro Glu Ser Glu 20 25 30

Val Ala Val Ala Met Gly Thr Ser Leu Gin lie Thr Cys Ser Met Ser 35 40 45

Cys Asp Glu Gly Val Ala Arg Val His Trp Arg Gly Leu Asp Thr Ser 50 55 60

Leu Gly Ser Val Gin Thr Leu Pro Gly Ser Ser lie Leu Ser Val Arg 65 70 75 80

Gly Met Leu Ser Asp Thr Gly Thr Pro Val Cys Val Gly Ser Cys Gly

85 90 95

Ser Arg Ser Phe Gin His Ser Val Lys lie Leu Val Tyr Ala Phe Pro 100 105 110

Asp Gin Leu Val Val Ser Pro Glu Phe Leu Val Pro Gly Gin Asp Gin 115 120 125

Val Val Ser Cys Thr Ala His Asn lie Trp Pro Ala Asp Pro Asn Ser 130 135 140

Leu Ser Phe Ala Leu Leu Leu Gly Glu Gin Arg Leu Glu Gly Ala Gin 145 150 155 160

Ala Leu Glu Pro Glu Gin Glu Glu Glu lie Gin Glu Ala Glu Gly Thr

165 170 175

Pro Leu Phe Arg Met Thr Gin Arg Trp Arg Leu Pro Ser Leu Gly Thr 180 185 190

Pro Ala Pro Pro Ala Leu His Cys Gin Val Thr Met Gin Leu Pro Lys 195 200 205

Leu Val Leu Thr His Arg Lys Glu lie Pro Val Leu Gin Ser Gin Thr 210 215 220

Ser Pro Lys Pro Pro Asn Thr Thr Ser Ala Glu Pro Tyr lie Leu Thr 225 230 235 240 Ser Ser Ser Thr Ala Glu Ala Val Ser Thr Gly Leu Asn lie Thr Thr

245 250 255

Leu Pro Ser Ala Pro Pro Tyr Pro Lys Leu Ser Pro Arg Thr Leu Ser 260 265 270

Ser Glu Gly Pro Cys Arg Pro Lys lie His Gin Asp Leu Glu Ala Gly 275 280 285

Trp Glu Leu Leu Cys Glu Ala Ser Cys Gly Pro Gly Val Thr Val Arg 290 295 300

Trp Thr Leu Ala Pro Gly Asp Leu Ala Thr Tyr His Lys Arg Glu Ala 305 310 315 320

Gly Ala Gin Ala Trp Leu Ser Val Leu Pro Pro Gly Pro Met Val Glu

325 330 335

Gly Trp Phe Gin Cys Arg Gin Asp Pro Gly Gly Gin Val Thr Asn Leu 340 345 350

Tyr Val Pro Gly Gin Val Thr Pro Asn Ser Ser Ser Thr Val Val Leu 355 360 365

Trp lie Gly Ser Leu Val Leu Gly Leu Leu Ala Leu Val Phe Leu Ala 370 375 380

Tyr Arg Leu Trp Lys Cys Tyr Arg Pro Gly Pro Arg Pro Asp Thr Ser 385 390 395 400

Ser Cys Thr His Leu

405

Claims

WHAT IS CLAIMED IS:

1. An isolated DNA sequence other than as a part of a chromosome encoding a mucosal addressin characterized by having three immunoglobulin-like domains, a mucin-like region, between the second and third domains, a transmembrane domain and a cytoplasmic domain, wherein said mucosal addressin is capable of binding to a leukocyte surface membrane protein, or single or double stranded fragment of said DNA of at least about 12 nt in length.

2. An isolated DNA according to Claim 1, wherein said DNA is cDNA.

3. An isolated DNA according to Claim 1, wherein said cDNA is mouse cDNA.

4. An isolated DNA according to Claim 1, wherein said fragment encodes at least one of said immunoglobulin like domains or said mucin-like domain.

5. An isolated DNA according to Claim 1, wherein said fragment encodes cDNA of a differentially spliced mucosal addressin.

6. An isolated DNA according to Claim 1 , wherein said fragment encodes a soluble form of said mucosal addressin including at least the extracellular portion of said mucosal addressin.

7. Purified MAdCAM-1 characterized by having three immunoglobulin-like domains, a mucin-like region, between the second and third domains, a transmembrane domain and a cytoplasmic domain and being capable of binding to a leukocyte surface membrane protein or fragment thereof of at least about 12 amino acids.

8. Purified MAdCAM-1 according to Claim 7, wherein said MAdCAM-1 is mouse MAdCAM-1.

9. Purified MAdCAM-1 according to Claim 7, wherein said MAdCAM-1 comprises a fragment of MAdCAM-1 comprising at least one immunoglobulin domain or said mucin region.

10. Purified MAdCAM-1 according to Claim 7, wherein said

MAdCAM-1 is the expression product of a genetic construct in an expression host.

11. Purified MAdCAM-1 according to Claim 10, wherein said host is a prokaryotic host and said MAdCAM-1 lacks glycosylation.

12. Purified MAdCAM-1 according to Claim 10, wherein said host is a eukaryotic host.

13. A purified fragment of MAdCAM-1, comprising the extracellular portion of MAdCAM-1, said fragment being soluble in blood.

14. A genetic construct comprising a DNA sequence according to Claim 1 under the transcriptional and translational control of transcriptional and translational initiation and termination regions functional in an expression host.

15. A genetic construct according to Claim 14 joined to a vector for introduction into an expression host, said vector comprising a marker for selection in said expression host.

16. A genetic construct according to Claim 15, wherein said host is a prokaryote.

17. A genetic construct according to Claim 15, wherein said host is a eukaryote.

18. A method for modulating homing of leukocytes to mucosal endothelial cells comprising MAdCAM-1, said method comprising: adding to a cell population comprising leukocytes capable of binding to said MAdCAM-1 and endothelial cells expressing MAdCAM-1, a modulating amount of MAdCAM-1 or a fragment of MAdCAM-1 capable of interfering with the binding of said leukocyte with said endothelial cell.

19. A conjugate comprising MAdCAM-1 or fragment thereof of at least 12 amino acids covalently joined to a moiety capable of providing a detectable signal, an immunogen, or an agent capable of producing a physiological reaction.

20. A conjugate of at least 12 amino acid fragment of MAdCAM-1 and at least one of: a hapten; an antigen; a compound capable of providing a detectable signal; a toxin or an antibody; or a fused protein comprising said at least 12 amino acid fragment.

21. A method for screening for molecules capable of binding

MAdCAM-1, said method comprising: combining MAdCAM-1 or functional fragment thereof with said molecule; and detecting the presence of complexes.

22. A method for screening for molecules capable of inhibiting binding of leukocytes to MAdCAM-1, said method comprising: combining leukocytes with MAdCAM-1 and said molecule under conditions whereby leukocytes bind to MAdCAM-1; and detecting the degree of binding of leukocytes with MAdCAM-1 as compared to the degree of binding of leukocytes to MAdCAM-1 in the absence of said molecule.

23. A method according to Claim 22, wherein said MAdCAM-1 is present in the cell membrane of a cell which does not normally express

MAdCAM-1.