MXPA97000200A - Molecula-1 of endothelial cellular adhesion deplaquetas, composiciones y meto - Google Patents

Abstract

The present invention provides novel isoforms substantially isolated from human platelet endothelial cell adhesion molecules-1, DNAs encoding transcripts encoding novel and other isoforms, including a previously identified soluble isoform, methods for using these DNAs to make the isoforms by the expression of the DNAs, and promoter segments, that control the transcription of genes from human platelet endothelial cell adhesion molecule-1. The novel isoforms differ from the full-length human platelet endothelial cell adhesion molecules-1, because they lack one or more segments near the C-terminus encoded by exons 10-15 of the gels for full-length molecules, and are presented in I live from the alternative junction of transcription from the gene

Description

MOLECULE-I OF ENDOTHELIC CELLULAR ADHESION OF PLATELETS, COMPOSITIONS AND METHODS TECHNICAL FIELD The present invention relates to a type of human protein molecule called "platelet endothelial cell adhesion molecule-1", and often referred to simply as "PECAM-1". It is known that platelet endothelial cell adhesion molecules occur on the surfaces of platelets and leukocytes in the blood, as well as on the surfaces of endothelial cells in the walls of blood vessels. Proteins are involved in the adhesion of these cell types to one another, and in the processes that involve this adhesion. As such, proteins are involved in different conditions that involve the blood system, such as inflammation associated with many injuries and diseases, atherosclerosis, and damage to blood vessels resulting from angioplasty. More particularly, the invention relates to certain novel modified forms, known as "isofor", of platelet endothelial cell adhesion molecules-1 (PECAM-1). It has been discovered in an unexpected way, that these isoforms determine the detailed structure and organization of the gene for molecules-1 of endothelial cell adhesion of platelets. The invention also relates to novel DNAs encoding the isoforms of the invention, expression vectors that can be used to make the isoforms of the invention, and novel promoter segments that control the in vivo expression of endothelial cell adhesion molecules-1. of platelets, from the genes that encode them.

BACKGROUND OF THE INVENTION Full-length mature platelet endothelial cell adhesion molecules-1 (PECAM-1) are glycosylated proteins with 711 amino acids and a molecular weight of approximately 130 kilodaltons. The proteins are members of the immunoglobusuperfamily. They are expressed on platelets, in the intercellular junctions of resting endothelial cells, and on circulating monocytes, granulocytes, and certain subsets of T cells. Newman et al. (I) (1990) Science 247, 1219-1222; Muller et al (I) (1989) J. Exp. Med. 170, 399-414; Albelda et al. (I) (1990) J. Cell. Biol. 110, 1227-1237; Ashman and Aylett (1991) Tissue Antigens 38, 208-212. From a molecular cloning study, it is known that platelet endothelial cell adhesion molecules-1 have 6 extracellular domains in the form of Ig, a short transmembrane region, and a cytoplasmic tail relatively long 118 amino acids (aa) containing multiple potential sites for phosphorylation, lipid modification, and other post-translational modifications. Newman et al. (I), supra, and Newman, United States Patent Number 5,264,554 (Patent * 554). Three full-length mature platelet endothelial cell adhesion molecules-1 have been found in the prior art. Newman, in the '554 patent, presented one of these, designated herein as "form 1", having the amino acid sequence shown in Figure 1 of the' 554 patent. Another, designated herein as "form 2", has the amino acid sequence described by Stockinger et al. (1990) J. Im unol. 145, 3889-3897. The third, designated herein as "form 3", has the amino acid sequence described by Zehnder et al. (1992) J. Biol. Chem. 267, 5243-5249. These references also present the DNA sequences encoding the respective forms. A fourth form ("form 4"), for which portions of the amino acid sequence and the sequence of the gene are provided in SEQ ID. NO: 3 to SEQ ID. NO: 11 hereinafter, has an asparagine, rather than a serine at the position of amino acid 536 (as numbered in Figure 1 of the '554 patent), due to a change from a 2-deoxyguanilate (G ) up to 2'- deoxyadenylate (A) at the position of nucleotide 1829, in the cDNA sequence of Figure 1 of the '554 patent (corresponding to the position of nucleotide 196 in SEQ ID NO: 3 hereinafter). The sequence reported by Stockinger et al. (1990), supra, also has an asparagine at amino acid position 536, because it has an A rather than a G at the position of cDNA 1829. With respect to the cDNA sequences shown in SEQ ID. NO: 3 to SEQ ID. NO: 11, there is an additional difference in the sequence of the cDNA shown in Figure 1 of the '554 patent: the base 108 in the 3 * sequence - not translated, of SEQ ID. NO: 11, is a 2'-deoxyguanilate, rather than a 2'-deoxyadenylate, as in Figure 1 of the '554 Patent. This base 108 corresponds to the base 2416 of the cDNA sequence of Figure 1 of the '554 Patent. In addition, at various nucleotide positions of the cDNAs for the platelet endothelial cell adhesion molecule-1, silent substitutions have been found (substitutions not resulting from amino acid changes). With reference to the cDNA sequence of Figure 1 of the '554 patent, this silent substitution has been found at position 1593 of the amino acid coding region, and as noted above, in the 3'-non-translated region, position 2416 (corresponding to the position of nucleotide 108 in SEQ ID NO: 11 hereinafter). See Newman et al. (I), supra; Stockinger et al., Supra; Zehnder et al., Supra; and the '554 Patent. Then, apparently, there are a number of almost identical alleles of the genomic DNA of molecule-1 of endothelial cell adhesion of platelets in the human genetic group. A polymorph of a molecule-1 of platelet endothelial cell adhesion with an amino acid substitution in the cytoplasmic domain, amino acids at positions 594-711, has not been found. A soluble form of a cell adhesion molecule-1 has been identified platelet endothelial, with a molecular weight between approximately 6,000 and 9,000 daltons less than that of a full length mature platelet endothelial cell adhesion molecule-1. Goldberger et al., Blo? D 80, 266a (1992). The platelet endothelial cell adhesion molecules-1 are important mediators of the cell-platelet-platelet-leukocyte-cell-platelet-endothelial cell interactions involved in the accumulation of platelets, in the development of the atherosclerotic plaque, and in the thrombus development as a result of vascular trauma, as may be caused, for example, by angioplasty or similar processes. The molecules-1 of endothelial cell adhesion of platelets are also involved in the interactions of leukocytes-endothelial cells involved in processes such as transendothelial cell migration and related phenomena, such as inflammation. Muller et al. (II) (1993) J. Exp. Med. 178, 449-460 and Vaporciyan et al. (1993) Science 262, 1580-1582, describe the use of platelet endothelial cell adhesion molecule-1 specific antibodies. , to interfere with neutrophil recruitment and transendothelial migration. The mechanisms by which the platelet endothelial cell adhesion molecules-1 mediate these cellular interactions are complex, since platelet endothelial cell adhesion molecules-1 can interact both homophilically (a molecule-1 of platelet endothelial cell adhesion that it is fixed to a molecule-1 of endothelial cell adhesion of platelets), Albelda et al. (II) (1991) J. Cell. Biol. 114, 1059-1068, as heterophilically (a molecule-1 of platelet endothelial cell adhesion that binds to a molecule other than a molecule-1 of platelet endothelial cell adhesion), Muller et al. (III) (1992) J. Exp. Med. 175, 1401-1404 and DeLisser et al. (1993) J. Biol.

Chem. 268, 16037-16046, to perform its adhesive functions. The cytoplasmic domain of a platelet endothelial cell adhesion molecule-1 is 118 amino acids with C-terminal, amino acids 594-711, in the mature molecule. This domain has an important role in the regulation of the adhesive properties of a molecule-1 of endothelial cell adhesion of platelets. It has been found that the removal of the C-terminal portions in the constructions of the endothelial cell adhesion molecule-1 of recombinant platelets, convert the molecule from a heterophilic to homophilic ligand binding specificity, DeLisser et al. (II) (1994). J. Cell. Biol. 124, 195-203. The cytoplasmic domain has sites for phosphorylation and lipid modification, and interacts with the cytoskeleton. Newman et al. (II) (1992) J. Cell. Biol. 119, 239-246 (1992); Zehnder et al., Supra. Modifications in the cytoplasmic domain affect not only the adhesive properties of an extracellular domain of platelet endothelial cell adhesion molecules-1, but also its subcellular distribution, interactions with intracellular signaling molecules, and the ability to participate in Intercellular signal transduction.

COMPENDIUM OF THE INVENTION The present invention is based on a study of the organization and structure of human genomic DNA, from the short arm of chromosome 17, from which a molecule-1 of platelet endothelial cell adhesion is expressed (" PECAM-1") human. In this study, it has been discovered that the gene for PECAM-1 includes 16 exons. In an unexpected way, it has been discovered that an unusually high number of exons are involved, 7 of them, exons 10 - 16, in the coding for the "cytoplasmic tail" of the PECAM-1 molecule, and that alternative forms are presented ("isoforms") of PECAM-1, with cytoplasmic tails that differ in the amino acid sequence, of the differential splicing of the transcription of the PECAM-1 gene. This differential splicing results in the exclusion, of the mRNA that is translated to make the protein, from the portions of the transcript corresponding to one or more of the exons 10-15. The resulting different mRNAs encode the PECAM-1 isoforms with different cytoplasmic tails. Accordingly, the invention provides substantially isolated isoforms of PECAM-1 that differ in their cytoplasmic tails. In this study, it has also been discovered that "P exon 9 of the PECAM-1 gene provides the part of the transcript that, if present in the mRNA, provides the domain of the protein that anchors it in the cell membrane. The transcription corresponding to exon 9 can also be excluded, alone or together with one or more of the transcripts corresponding to exons 10-15, by differential splicing from the mRNA encoding an isoform of a PECAM-1. has discovered as part of this invention, that the soluble PECAM-1 reported by Goldberger and collaborators, supra, results from an mRNA lacking the transcription encoded by exon 9. Other isoforms of the invention can be soluble, because they lack an amino acid segment that is required to anchor PECAM-1 or its isoforms in the cell membranes. These other isoforms include those that are encoded from an mRNA that lacks not only a part of the transcription of the PECAM-1 gene corresponding to one or more of the exons 10-15 of the gene, but also the part of the transcript corresponding to exon 9, or at least bases 44-100 of that exon, as shown in SEQ ID. NO: 4 later. Isoforms corresponding to the lack in the mRNA coding for the isoforms of a part of the transcription corresponding to one or more of exons 10-15 of the gene and exon 9 of the gene, can be presented from the differential splice. The isoforms corresponding to the lack in the mRNA that encodes the isoforms of a part of the transcription corresponding to one or more of the exons 10-15 of the gene and the bases 44-100, as shown in SEQ ID. NO: 4, exon 9 of the gene, can, like all isoforms of the invention, be made by expressing a cDNA with a sequence that is transcribed in an mRNA encoding the isoform. In addition, in the study underlying the invention, segments of DNA ("promoter segments") have been discovered that they act as promoters for the initiation and control of gene transcription for PECAM-1. These promoter segments allow the transcription of any DNA, with which they are operably linked for the initiation of transcription, substantially only in the cells of the vasculature, and particularly, the vascular endothelial cells, the leukocytes, or the platelet precursors (eg. example, megakaryocytes) where PECAM-1 is normally expressed. Accordingly, the promoter segments are useful for limiting to these cell types, the expression of a gene, to produce a protein of interest (possibly including a full-length PECAM-1 itself), or the transcription of a DNA to produce a Anti-sense RNA of interest. Other aspects of the invention, described more fully below in the specification and in the claims, flow from the discovery of the isoforms of the invention, the DNA segments encoding the transcripts encoding the isoform-, and lo :: r. ogmontor. promoters of the invention.

BRIEF DESCRIPTION OF THE SEQUENCES PROVIDED IN THE SEQUENCE LIST SEQ ID. NO: l is the sequence of the first (that is, the 5 '-more "upstream") 492 base pairs determined in the study underlying the present invention, of the gene for a human platelet endothelial cell adhesion molecule-1. The 492 base pairs immediately preceded that were determined in the study as the first base pairs most frequently used (ie, the 5 '- plus "upstream") of the first exon of the gene. The DNA segment with the sequence of SEQ ID. N0: 1 is a promoter segment of the invention. SEQ ID. NO: 2 is the sequence of base pairs 259-492 from SEQ ID. NO: l. Several subsegments of the segment with the sequence of SEQ ID. NO: 2 have characteristic sequences of the cis-acting elements that occur in the promoters. Accordingly, base pairs 1-7 of SEQ ID. NO: 2 have the sequence of a segment that occurs in the promoters associated with acute phase reagents; the base pairs 14-22 of SEQ ID. NO: 2 have the sequence of an inverted NF-kB site, which occurs in the promoters, whose transcription activity is regulated by the cytosines; the pairs of bnsor. 46-52 and 187-191 of SEQ ID. NO: 2 have 725 sequences of ets sites, which are recognized by the polyomavirus enhancer protein; the base pairs 207-216 have a sequence of an ets site combined with a consensus sequence of a GATA element (5 '-AGATA), which is known to be involved in the regulation of expression genetics in cells of the megakaryocytic line; and base pairs 200-223 have a consensus sequence of transcription initiator RNA polymerase II similar to that found in other promoters that, as the segment with the sequence of SEQ ID. NO: 2, lack the 5'TATA recognition sequence for that polymerase. The segment with the sequence of SEQ ID. NO: 2 is a subsegment that has the transcription control activity of the segment with the sequence of SEQ ID. NO: l. SEQ ID. NO: 3 is the sequence of exon 8, together with the sequence of the first (ie, the 5 '-more upstream ") twenty bases of the intron that follows immediately (ie, is 3' -from or" current down "from) exon 8 in the gene examined in the study underlying the present invention SEQ ID NO: 4 is the sequence of exon 9, together with the sequence of the latter (ie, the 3 '- more "downstream") twenty bases of the intron immediately preceding (ie, it is 5 * -de or "upstream" of) exon 9, and the sequence of the first twenty bases of the intron that immediately follows exon 9 In the gene examined in the study underlying the present invention, exon 9 encodes the segment of a platelet endothelial cell adhesion molecule-1, or an isoform thereof, which is called the "transmembrane domain". this isoform from a DNA segment, from which a segment of exon 9 comprising the exon from base 44 to base 100 has been deleted, as shown in SEQ ID. NO: 4 results in a soluble polypeptide, rather than a membrane-bound polypeptide. The amino acids 1-8 and 28-36 shown in SEQ ID. NO: 4, it is believed that they are involved in anchoring the polypeptide to the cell or to the platelet membrane, but, unlike amino acids 9 -. 9-27, need not be absent from a polypeptide, for the polypeptide to be soluble. The amino acids 9-27 shown in SEQ ID. NO: 4 correspond to amino acids 575-593 of the platelet endothelial cell adhesion molecules-1 for which the sequence is shown in Figure 1 of U.S. Patent Number 5,264,554 (the '554 Patent). SEQ ID. NO: 5 is the sequence of exon 10, together with the sequence of the last twenty bases of the intron that immediately precedes exon 10, and the sequence of the first twenty bases of the intron that immediately follows exon 10 in the gene examined in the study that underlies the present invention. SEQ ID. NO: 6 is the sequence of exon 11, together with the sequence of the last twenty bases of the intron that immediately precedes exon 11, and the sequence of the first twenty bases of the intron that immediately follows the exon 11 in the gene examined in the study underlying the present invention. SEQ ID NO: 7 is the sequence of exon 12, together with the sequence of the last twenty bases of the intron that immediately precedes exon 12, and the sequence of the first twenty bases of the intron that immediately follows exon 12 in the gene examined in the study underlying the present invention. SEQ ID NO: 8 is the sequence of exon 13, together with the sequence of the last twenty bases of the intron that immediately precedes exon 13, and the sequence of the first twenty bases of the intron that immediately follows exon 13 in the gene examined in the study underlying the present invention. SEQ ID NO: 9 is the sequence of exon 14, together with the sequence of the last twenty bases of the intron that immediately precedes exon 10 / and the sequence of the first twenty bases of the intron that immediately follows exon 14 in the gene examined in the study underlying the present invention. SEQ ID NO: 10 is the sequence of exon 15, together with the sequence of the last twenty bases of the intron that immediately precedes exon 15, and the sequence of the first twenty bases of the intron that immediately follows exon 15 in the gene examined in the study that underlies the present invention. SEQ ID NO: 11 is the sequence of the 3 • -more (plus "downstream") 921 bases determined in the study underlying the present invention, of the gene for a human platelet endothelial cell adhesion molecule-1. The sequence of the first twenty bases in SEQ ID. NO: 11 is the sequence of the last twenty bases of the intron that immediately precedes exon 16 of that gene. The sequence of the last 901 bases in SEQ ID NO: 11, is the sequence of the first (5 '-more) 91 base pairs of that exon 16. Of these 901 base pairs, 871 are at 3' -del triplet corresponding to the stop codon in the mRNA that encodes the full-length protein and any isoform thereof, for which the reading frame of the mRNA is not altered due to the lack of an exon. There is no polyadenylation signal sequence in 5'-AATAAA primary consensus among these 871 base pairs. However, two polyadenylation signal sequences in secondary consensus are presented in bases 380-385, and in bases 536-541 of SEQ II) NO: 11. base 249 of SEQ ID NO: 11 corresponds to the base 2557 of CDNA encoding the platelet endothelial cell adhesion molecule provided in Figure 1 of the '554 patent. SEQ ID NO: 3 to SEQ ID NO: 11 also show the amino acid sequences encoded by the portions of these sequences corresponding to the exons. In the cases where an intron interrupts the triplet corresponding to a codon for an amino acid, the encoded amino acid is shown with the sequence of the exon that includes two of the three bases of the triplet. SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14 and SEQ ID NO: 15 are the sequences of the primers used in the study underlying the present invention. SEQ ID NO: 12 is complementary to the sequence of bases 100-141 of the cDNA encoding the platelet ondotelial cell adhesion molecule provided in Figure 1 of the '554 patent. SEQ ID NO: 13 is complementary to the sequence of bases 2465-2482 of that cDNA sequence of Figure 1 of the '554 patent. SEQ ID NO: 14 has the same sequence as bases 2085-2102 of that cDNA sequence of Figure 1 of the '554 Patent. Finally, SEQ ID NO: 15 has the complementary sequence for that of bases 2324-24343 of that cDNA sequence of Figure 1 of Patent '554. SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20 and SEQ ID NO: 21 are sequences of primers used to construct, from the cDNAs encode isoforms of the invention having a segment corresponding to bases 44-100 of exon 9, as shown in SEQ ID NO: 4, cDNAs lacking that segment, and therefore, which can be used to make isoforms soluble of the invention. SEQ ID NO: 16 is the sequence of bases 1602-1621 of the cDNA, for which the sequence is provided in Figure 1 of the '554 Patent. The sequence of the 16 bases 5 '-more than SEQ ID NO: 17, is the sequence complementary to that of bases 1929 -1944 in the cDNA, for which the sequence is provided in Figure 1 of the' 554 Patent , and the sequence of the 18 bases 3 '-more than SEQ ID NO: 17, is the complementary sequence for that of the bases 2002-2019 of that cDNA. SEQ ID NO: 18 is complementary to SEQ ID NO: 17. SEQ ID NO: 19 is the complementary sequence for that of bases 2465 -2482, SEQ ID NO: 20 is the same as that of bases 1754 - 1773, and SEQ ID NO: 21 is the complementary sequence for that of bases 2324-24343 in the cDNA, for which the sequence is provided in Figure 1 of the '554 Patent.

DETAILED DESCRIPTION OF THE INVENTION In one of its aspects, the invention is a substantially isolated isoform of a human platelet endothelial cell adhesion molecule-1, which is full-length and mature, wherein the isoform has the resulting amino acid sequence of the translation of the mRNA which is the same as an mRNA that is transferred to make the human platelet endothelial cell adhesion molecule-1, except that the mRNA that is transferred to make the isoform, lacks one or more of the segments of MRNA corresponding to exons 10 - 15 of the gene for the human platelet endothelial cell adhesion molecule-1, and optionally also lacks the mRNA segment corresponding to exon 9 in its entirety of said gene, or any continuous segment of exon 9 comprising the base pairs 44 - 100, as shown in SEQ ID NO: 4, and has a number of base pairs that is uniformly divisible by 3. In another of its aspects, the invention is a segment of DNA that encodes that for a transcription for a isoform of a human platelet endothelial cell adhesion molecule-1, which is full length, wherein the isoform has the amino acid sequence that results from the translation of the mRNA which is the same as an mRNA that is transferred to make the molecule -1 of human platelet endothelial cell adhesion, except that the mRNA that is transferred to make the isoform, lacks one or more of the mRNA segments corresponding to exons 9-15 of the gene for the mole human platelet endothelial cell adhesion cell-1, and if this mRNA that is transferred to make the isoform does not lack the entire segment corresponding to exon 9 of said gene, the mRNA that is transferred to make the isoform optionally lacks the segment of MRNA corresponding to any continuous subsegment of exon 9 comprising base pairs 44-100, as shown in SEQ ID NO: 4, and has a number of base pairs that is uniformly divisible by 3. The DNA segments of the invention can be included in expression vectors operably for the expression of isoforms, for which the DNA segments of the invention encode transcripts, in cells, in culture (including cultures maintained in vivo in animals, such as in the peritoneal cavities of mice or rats), or in vivo in humans. Preferred expression vectors would provide for the expression of isoforms of the invention in mammalian cells in culture, or in cells of the vasculature in vivo in humans. A promoter segment of the invention (see below) can be employed to drive expression specifically in human vasculature cells. Accordingly, a DNA segment of the invention can be used to make, in cultured cells or otherwise, as indicated for expression vectors, the isoform, for which the segment codes for a transcript, in a method comprising the expression of the DNA segment; and the invention encompasses these methods of using the DNA segments of the invention. The invention also encompasses expression vectors, cells transformed with those vectors, cultures of those cells, and methods for making isoforms of the invention by expression, in cultured cells or otherwise, from expression vectors of the invention. Still further, the invention involves a promoter segment that (i) has a sequence that is substantially the same as SEQ ID NO: 1, or any subsequence thereof, such as, for example, SEQ ID NO: 2, which is active in the control of transcription with respect to (ie, capable of initiating or controlling the transcription of) a segment of DNA operably linked to it for transcription, and (ii) is not on the long arm of a human chromosome 17 if the promoter segment is immediately placed 5 'for the gene for a human platelet endothelial cell adhesion molecule-1. The invention still further involves the method of using a promoter segment of the invention to control the transcription of a DNA segment to make an antisense RNA, or to control the expression of a gene to make a protein (possibly including, but not limited to) a, full-length PECAM), in cells of the vasculature, especially vascular endothelial cells, leukocytes, and platelet precursors, for example, megakaryocytes, where PECAM-1 is naturally expressed. "Substantially isolated", with respect to the isoforms of the invention, means "separated from the environment in which the isoform occurs in nature". Therefore, among other possibilities, the isoforms in cells in culture, the isoforms in living humans, in cells where the isoforms do not occur naturally or occur naturally, but at a different concentration, the isoforms on chromatographic or electrophoretic gels, in liposomes, or in solution, would be "substantially isolated. " Certainly, isoforms in aqueous solution, such as a solution that is pharmaceutically acceptable for administration to a human being by injection or otherwise, would be "substantially isolated". In these pharmaceutically acceptable solutions, an isoform of the invention may be present in a concentration between about 10 nM and the lowest of about 50 μM, and the solubility of the isoform in the solution. A concentration of at least about 50 nM, and more usually at least about 1 μM, will be used. Conveniently, an isoform of the invention including the transmembrane domain (the amino acid segment at positions 575-593 of Figure 1 of the '554 Patent, which corresponds to amino acids 9-27 of SEQ ID NO: 4) is incorporated. , in a liposome for suspension in solution and administration to a human being, typically by infusion or intravenous or intraarterial injection of this solution. The reference to a "mature" isoform or other protein, means that the protein does not have a signal peptide. The reference to a "full-length" platelet endothelial cell adhesion molecule-1, means one having the entire amino acid sequence naturally occurring in the molecule. In the case of the four human forms of these molecules that are now known, after the present invention, this "full length" sequence has 711 amino acids in the "mature" molecules, and 738 amino acids in the "full length" molecules with the signal peptide. A "cDNA segment" that codes for an RNA, means that the DNA segment can be transcribed to make the RNA, but it does not have introns. Unless otherwise qualified, a "cDNA" or a "cDNA segment" is double stranded. A cDNA segment for a protein or a polypeptide means a segment of DNA that has no introns, and that can be transcribed into an RNA that, in turn, can be translated to make the protein or the polypeptide (I say, " encodes "the protein or the polypeptide. A "gene" for a protein or a polypeptide means: (1) a segment of a genomic DNA for the protein or polypeptide, whose segment is transcribed in the RNA, but may have introns, where the RNA, after being processed for remove the segments corresponding to the introns, is capable of moving into the protein or polypeptide, or (2) a segment that is not a cDNA segment, but differs from that specified in part (1) of this paragraph only by one or more nucleotide substitutions, deletions , and additions that, taken together, are silent. These substitutions, deletions and additions are "silent" altogether, if the segment with them is transcribed into an RNA that, after being processed to remove the segments corresponding to the introns, is able to move inside the same protein or polypeptide. A "genomic DNA" means a DNA that is part of a chromosome, and can include not only a "gene" (a segment that is transcribed), but also segments that are not transcribed, such as promoter segments, that control the transcription of the genes that can be part of DNA. All the amino acids referred to in this specification, except for the non-enantiomorphic glycine, are L-amino acids. An amino acid can be referred to by using the conventional three-letter designation, as indicated in the following Table I.

TABLE I Designations for Amino Acids Amino Acid Designation of Three Letters L-alanine Ala L-arginine Arg L-asparagine Asn L-aspartic acid Asp L-cysteine Cys L-glutamic acid Glu L-glutamine Gln glycine Gly L-histidine His L-isoleucine lie L-leucine Leu L-lysine Lys L-methionine Met L- phenylalanine Phe L-proline Pro L-serine Ser L-threonine Thr L-tryptophan Trp L-tyrosine Tyr L-Valine Val Peptide or polypeptide sequences are written and numbered from the amino-terminal amino acid to the carboxy-terminal amino acid. One-letter conventional codes, "A", "C", "G" and "T" are used in the present for nucleotides adenylate, cytidylate, guanylate and thymidylate, respectively. The skilled person will understand that, in the DNAs, the nucleotides are 2'-deoxyribonucleotide-5'-phosphates (or, at the 5 'end, triphosphates), whereas, in the RNAs, the nucleotides are ribonucleotide-5'-phosphates ( or, at the 5 'end, triphosphates), and uridylate (U) occurs in place of T. "N" means any of the four nucleotides. "dNTP" means any of the four 2'-deoxyribonucleoside-5'-triphosphates. Oligonucleotide or polynucleotide sequences are written from the 5 'end to the 3' end. A promoter segment is "substantially the same" as one of the specified sequence, if its sequence differs in one or more positions from that of the segment of the specified sequence but, in a mammalian cell of the vasculature, such as an endothelial cell of the human umbilical vein ("HUVEC", Newman et al. (III) (1986) J. Cell. Biol. 103, 81-86), or an ECV304 cell (ATCC CRL-1998), segment initiates transcription at an index that is at least 10 percent, and more preferably at least 50 percent of the transcription initiation index per segment of the specified sequence, in a conventional assay for promoter activity. See, for example, Sections 9.6 and 9.7 of Current Protocols in Molecular Biology, edited by Ausubel et al.

Current Protocols, a joint risk between Greeno Publishing Associates, Inc. and John Wiley & Sons, Inc. (1994). The placement of a promoter segment "operably for transcription" with respect to a segment that is to be transcribed, is understood in a direct way by a person of ordinary experience. It means orienting and placing the promoter segment in a larger DNA segment comprising both the promoter segment and the segment to be transcribed under the control of the promoter segment, so that, in a cell where the promoter segment is effective to initiate transcription, the segment to be transcribed is in fact transcribed in a transcription process that starts from the promoter segment. It would be routine for a skilled person to determine, by methods well known in the art, sub-segments of a promoter segment of the invention, which remain active in the control of transcription. The subsegments are prepared by any of numerous conventional techniques, for example, Chapter 8 of Current Protocols in Molecular Biology, supra. The test for the transcription control activity of a subsegment can be any conventional assay for promoter activity, as indicated above, in Sections 9.6 and 9.7 of Current Protocols in Molecular Biology, supra. The segments and sub-segments promoting the invention can be operably linked for transcription to genes, or preferably cDNA, so that proteins of interest are expressed in the cells of the vasculature. These proteins can include not only PECAM-1 or isoforms thereof, but also proteins, such as adenosine deaminase, to treat the immune deficiency due to deficiency of that enzyme, or Factor IX, to treat the form of hemophilia due to the lack of that factor. The resulting constructs can then be transformed into vasculature cells, such as endothelial cells, leukocytes, or platelet precursor cells, by conventional transformation techniques, including, without limitation, retroviral mediated transformation. The transformed cells, in the person to be treated, will then produce the protein of interest. The promoter segments and subsegments of the invention can be operably linked for transcription to the DNAs that are oriented with respect to the direction of transcription from the promoter segment or subsegment to produce, on transcription, anti-sense RNAs in cells of the vasculature As understood by the experts, this anti-sense RNA, by hybridizing in the cell in which it becomes a segment of DNA or RNA that is complementary in sequence, blocks the function of the DNA or RNA segment, and this way eliminates or inhibits the expression of a gene that depends on that function. For example, an anti-sense RNA could inhibit the expression of a gene for a protein by hybridizing all or part of the mRNA that moves into the protein, thereby blocking the translation of the mRNA. A promoter segment, or a subsegment thereof, of the invention, could be used to transcribe a cDNA, for all or part of a PECAM-1, whose cDNA is oriented, with respect to the direction of transcription from the promoter segment or subsegment. , to produce an RNA with the sequence complementary to that of all or part, respectively, of the mRNA that is transferred to produce the PECAM-1. The amount of PECAM-1 produced by the cell thus transformed would be reduced or eliminated. This would be of therapeutic advantage in the treatment of conditions, such as inflammation, dependent on the transendothelial migration of leukocytes, or arterial occlusion, dependent on the fixation between platelets and cells that naturally have PECAM-1 on their surfaces, where they are involved the interactions of PECAM-1 molecules with other PECAM-1 molecules or with other cell surface proteins. To make the isoforms of the invention, reference is made to the description of the '554 Patent. The isoforms that are encoded by the mRNAs that lack all the segments corresponding to one or more exons, different from the exon 9, occur naturally, at low levels, in the membranes of the cells of the vasculature, and thus, can be isolated from these cells by direct modifications of the procedures described in Patent '554, to obtain the PECAM- l from the cells in a recoverable form. In these methods for isolating isoforms from cells, antibodies that are obtained using as antigen PECAM-1 isoforms produced in large quantities from the cells in culture transformed with expression vectors to make the isoforms can be used. The isoforms of the invention that lack the amino acid segment corresponding to exon 9, or of the amino acid segment corresponding to amino acids 9-27 of that segment corresponding to exon 9, as shown in SEQ ID NO: 4, are soluble. Those lacking the segment corresponding to exon 9, occur naturally in the blood serum, and can be isolated from it by conventional techniques, again using alternatively antibodies that can be developed using as an antigen the corresponding PECAM-l isoforms, respectively , produced in large quantities from cells in culture transformed with expression vectors to make the isoforms. The isoforms of the invention, and the isoforms that they lack the entire amino acid segment corresponding to exon 9, preferably they are made by using segments of DNA of the invention as part of expression vectors, and by culturing cells that have been transformed with the expression vectors to make the isoforms by the expression of the DNAs encoding the isoform of the invention. In this regard, with respect to the preparation of expression vectors for making the isoforms in bacteria, yeast or mammalian cells, reference can be made not only to the '554 patent, but also to the standard works related to those vectors, including , for example, Current Protocols in Molecular Biology, supra. It would be preferred to prepare the isoforms in mammalian cells, and particularly human cells, in culture, in such a way that their glycosylation is similar to that of the naturally occurring forms. For expression in mammalian cells, cells developed using a system involving amplification of the copy number of the gene of interest are preferred, using a dihydrofolate / reductase-methotrexate system. See Chapter 16, and especially Section 16.14 of Current Protocols in Molecular Biology, supra. The isoforms of the invention can be used as described for PECAM-1 in the '554 patent. Accordingly, the isoforms of the invention are they can be used, for example, to make antibodies, and preferably monoclonal antibodies, for different diagnostic and therapeutic uses. The isoforms of the invention are to be administered to humans under the guidance of a physician. The soluble isoforms of the invention, or provided with the DNA of the invention, can be administered to humans in a single dose, by a single intravenous or intraarterial injection, or by intravenous or intra-arterial infusion, for a period of time between approximately 1 minute and 10 hours, in a water vehicle that is pharmaceutically acceptable, of between about 0.05 milligrams / kilogram and about 5 milligrams / kilogram of body weight of the recipient to alleviate the inflammation due to leukocyte transmigration that occurs in many situations, including arthritis, bee stings, spider bites, sepsis, anaphylactic concussion, and other conditions, or to inhibit arterial occlusions associated with atherosclerosis or vascular trauma, due to angioplasty or the like. The actual dosage regimen will vary from individual to individual, depending on, among other factors, the purpose for which an isoform is being administered, and the individual's medical condition. The isoforms of the invention that are not soluble, they can be combined with liposomes, and the resulting liposomes can be administered to have effects similar to those of the soluble isoforms. The invention will now be described in further detail in the following examples, which are provided for purposes of illustrating and demonstrating the invention, but not to limit it.

EXAMPLE 1 This Example provides a description of a study of the structure and organization of genomic DNA for a PECAM-1, and a determination of cDNA sequences that encode mRNA that encodes different isoforms of PECAM-1. Two human genomic libraries, constructed from DNA from peripheral blood leukocytes partially digested with Sau3AI cloned in lambda phage vectors, were screened and produced most of the PECAM-1 gene. The first library, in? EMBL3, was obtained from Clontech Laboratories (Palo Alto, California, USA). The second library in? GEM-11 came from Novagen (Madison, Wisconsin, USA). The screening of the library was performed by plate-raising hybridization (Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York), using full length or partial length PECAM-1 cDNA probes labeled with 32P protein, by random priming with [a-32P] dCTP (DuPont, New England Nuclear, Boston , Massachusetts, USA), and an oligo-labeling kit (Pharmacia, Piscataway, New Jersey, USA) (Feinberg and Vogelstein (1983) Anal. Biochem., 132, 6-13). Positive clones were purified from plaques, and phage DNA was isolated following conventional procedures (Sambrook et al., Supra), for the characterization of the genomic insert. An approximately 45 kb genomic clone containing a portion of the PECAM-1 genomic DNA was derived from a phagemid Pl library (clone # 530, Genome Systems, St. Louis, Missouri, USA), by reaction-based screening. of polymerase chain, using specific primers of PECAM-l. All inserts were characterized by restriction endonuclease mapping. Restriction fragments containing exons were identified by Southern blot hybridization of restriction endonuclease digestions with PECAM cDNA probes labeled with 32 P. Restriction endonuclease digestions of the genomic clone DNA, separated by 1 percent agarose gel electrophoresis, were transferred to nylon membranes (Boehringer Mannheim Biochemicals, Indianapolis, Indiana, USA, or Micron Separations, Inc., Westborough, Massachusetts, USA) by vacuum transfer. Prehybridization and hybridization were performed at 68 ° C in 5x Denhardt's solution (Sambrook et al., Supra), 6x SSC (lx SSC: 0.15 M sodium chloride, 0.015 M sodium citrate, pH 7), ethylenediaminetetraacetic acid 5 mM disodium (EDTA), 10 mM sodium phosphate, pH of 7, 1 percent sodium dodecyl sulfate (SDS), 50 micrograms / milliliter of herring sperm DNA. The labeled probe was added at 2-3 x 106 cpm / ml, and allowed to hybridize for 18 hours. The membranes were washed at 68 ° C: twice in 2x SSC, 0.1% SDS, once in 0.5x SSC, 0.1% SDS, and twice in O.lx SSC, 0.1% SDS; and each wash lasted 30 minutes. In certain cases, the exons were too small to be detected by hybridization of double-stranded DNA probes. Accordingly, when appropriate, hybridization of the oligonucleotide was performed. Synthetic oligonucleotides derived from the cDNA sequence of PECAM-1 (see the '554 patent) were labeled with 32 P, with [? -32P] dATP (DuPont, New England Nuclear), and T4 polynucleotide kinase (Promega, Madison, Wisconsin, USA, or New England Biolabs, Boston, Massachusetts, USA). The membranes were prehybridized in a solution of Denhardt 5x, 6x SSC, EDTA M, 10 mM sodium phosphate, pH of 7, SDS at 1 percent. Hybridizations were performed at 50 ° C for at least 18 hours in 10X Denhardt's solution, 5x SSC, 5mM EDTA, 20mM sodium phosphate, pH 7, 7 percent SDS, 100 microgram / milliliter DNA herring sperm. The membranes were washed at 50 ° C, three times, 20 minutes each, in a solution of Denhardt lOx, 3x SSC, 70 mM sodium phosphate, 5 percent SDS, then once or twice at 60 ° C in lx SSC, 1 percent SDS. All the hybridization procedures were done using a Model 310 hybridization oven (Robbins Scientific, Sunnyvale, California, USA). Each genomic insert was subcloned as fragments into plasmid vectors, ptzldr (Stratagene, La Jolla, California, USA), or pGEM-7 (Promega) for another gene mapping and direct sequence analysis. Most of the gene sequence - all exons, exon / intron junctions, and most of the intron sequence - was obtained using T7 DNA polymerase (Sequenase® brand, United States Biochemical, Cleveland, Ohio, USA) ), and [? -35S] dATP (DuPont, New England Nuclear). Several intron sequences were obtained by cycle sequencing using Taq polymerase, the Prism Dye Deoxyterminator kit, and the automated sequencing apparatus ABI 373A (Applied Biosystems, Foster City, California, USA). All sequences were determined according to the dideoxynucleotide termination method of Sanger et al. (1977), Proc. Nati Acad. Sci. (USA) 74, 5463-5467 (the Sanger method). Intron distances were determined by a combination of restriction mapping, amplification with polymerase chain reaction, and direct sequencing procedures. For genomic Southern blot hybridization, human genomic DNA was isolated from peripheral blood leukocytes separated from 50 milliliters of human blood taken from a normal healthy volunteer (see Poncz et al. (1982) He oglobin 6, 27-36). Ten micrograms of human genomic DNA with different restriction endonucleases were digested for 18 hours at 37 ° C. The digestions were separated through a 0.8 percent agarose gel by electrophoresis in a pH regulator of Tris-borate-EDTA at 30 volts for 18 hours, transferred to a nylon membrane, and hybridized with? DNc of PECAM-l labeled with 32P at 65 ° C for at least 18 hours in a 5x Denhardt, 6x SSC, 5 mM EDTA, 10 mM sodium phosphate, pH 7, 1 percent SDS, 100 micrograms / milliliter of herring sperm DNA. The membranes were washed at 65 ° C twice with 2x SSC, 0.1% SDS, then twice with lx SSC, 0.1% SDS; and each wash lasted 30 minutes. The Washed membranes were exposed to a Kodak XOMatAR or XRP film (Fotodyne, Milwaukee, Wisconsin, USA) for one to two weeks in the presence of an amplification screen. The polymerase chain reaction was carried out as follows. One microgram of total human genomic DNA, 20 nanograms of lambda phage DNA, or 2 nanograms of plasmid DNA, were routinely used as the starting material for 100 microliter polymerase chain reaction amplifications. The polymerase chain reactions were performed in 10 mM Tris-HCl, pH 8, 1.5 mM MgCl 2, 50 mM KCl, 0.01 percent gelatin, and 0.2 mM each of dNTP (2'-deoxyribonucleoside triphosphate). Primers were added to a final concentration of 0.5 μM. Amplification by polymerase chain reaction was performed in a thermal cycler (MJ Research, Inc. Watertown, Massachusetts, USA), using the following protocol: (1) 3 minutes of DNA denaturation at 100 ° C, (2) 2 minutes tempering of the initial primer at 55-57 ° C, (3) heating to 72 ° C followed by addition of a Taq poly erasa unit (Perkin-Elmer Corp. Oakbrook, Illinois, USA), (4) 1 to 5 minutes extension at 72 ° C, (5) 1.0 minutes of denaturation at 96 ° C, (6) 1.0 minutes of priming the primer at 55-57 ° C, (7) 30 cycles of steps 4 to 6, (8) 7 final minutes of extension, and (9) cooling to 4 ° C. The extension of the primer and the analyzes of Rapid Amplification-5 »of cDNA ends (RACE) were performed as follows. Umbilical cord umbilical cord endothelial cells were harvested, and primary cultures were established as described by Newman et al. (III) (1986) J. Cell. Biol. 103, 81-86. Total RNA and PoliA mRNA were isolated, according to the previously published methods (Lyman et al. (1990) Blood 75, 2343-2348). Yeast tRNA was obtained from Life Technologies, Inc. (Gaithersburg, Maryland, USA). Primer extension reactions were conducted using the oligonucleotide of SEQ ID NO: 12. The primer was labeled with [α -3 P] ATP by kinase, and hybridized with either 15 micrograms of total RNA, 4 micrograms of RNA of polyA, or 15 micrograms of yeast tRNA. Hybridization reactions were conducted at 56 ° C in 125 mM Tris, pH 8.3, containing 190 mM KCl, MgCl27.5 mM. Extension reactions were performed at 40 ° C in 50 mM Tris buffer, pH 8.3, containing 75 mM KCl, 10 mM dithiothreitol (DTT), 3 mM MgCl 2. 0.5 mM each of dNTP, 0.05 milligrams / milliliter of actinomycin D, 0.1 U / ml of RNasin® brand ribonuclease inhibitor (Promega), and 0.02 U / ml of MMLV reverse transcriptase. After extension, the unprotected RNA was digested with RNaseH (10 U / ml), followed by ethanol precipitation of the intact RNA-DNA hybrids. The products were electrophoresed through a polyacrylic amide gel denaturant at 5 percent, and visualized by autoradiography. The sequencing reactions using the above primer and a genomic clone containing the 5 'end of the PECAM-1 gene (see below), were conducted using the Sanger method. The 5 '-RACE reactions were conducted according to the manufacturer's instructions (Clontech, Inc.). The identification of splice variants of PECAM-1 mRNA was made as follows. Total RNA was isolated from endothelial cells of the human umbilical vein by the method of Chomczynski and Sacchi (1987) Anal. Biochem. 162, 156-159 (1987). The cDNA was generated from the RNA isolated with the anti-sense primer of SEQ ID NO: 13 by reverse transcription at 37 ° C, using MMLV reverse transcriptase (Boehringer Mannheim Biochemicals). The polymerase chain reaction amplification of the cDNA employed a forward primer in exon 11 with the sequence SEQ ID NO: 14, and an anti-sense primer that extended the binding of exons 15/16 having the SEQ ID sequence. NO: 15. The products of the polymerase chain reaction were separated by electrophoresis in 2 percent agarose gel. Occasionally, products of the polymerase chain reaction were separated directly from the gel, subcloned into the plasmid vector pGEM-5 (Promega), and sequenced as described above.

Using a chromosomal localization analysis of human / hamster somatic cell hybrid clones, it was discovered that genomic DNA for PECAM-1 (all forms) is presented in a copy on the long arm of human chromosome 17. In order to To determine the organization of the genomic DNA for human PECAM-l, two different lambda phage libraries and a phageid Pl library were screened using a combination of polymerase chain reaction amplification (Pl phagemid library), and hybridization with specific probes from PECAM-1 (phage libraries), as indicated above. A total of six genomic clones were obtained, with inserts that averaged approximately 15 kb (kilopars of bases) in size. The initial restriction mapping of these six clones revealed a genomic DNA of approximately 65 kb in size. The nucleotide sequence was determined for 30,127 base pairs of the PECAM-1 genomic DNA, including 561 bp (base pairs) 5 'for C at position 7 (the first base that is not part of the 5' EcoRI site). -GAATTC artificial) at the 5 'end of the cDNA sequence of PECAM-l previously reported (see Figure 1 of the' 554 patent). In order to locate the 5 'end of the genomic DNA encoding transcription of PECAM-1 mRNA, the principle of exon 1) within this 561 bp segment, the analysis of primer extension was conducted. An anti-sense oligonucleotide with SEQ ID NO: 12, a sequence complementary to that of a segment in the 5 * region of PECAM-1 mRNA, was used to prime the reverse transcription of endothelial cells of the human umbilical vein (HUVEC ), lung carcinoma cells A549, and yeast RNAs. (Lung carcinoma cells and yeast cells do not express PECAM-1). A single specific band was obtained for the HUVEC mRNA corresponding to A at the nucleotide position immediately 3 'to nucleotide 492 in SEQ ID NO: 1. This nucleotide is 204 bp upstream from the translation start site reported in Figure 1 of the '554 Patent. The products of the 5'-RACE polymerase chain reaction, derived from the 5 'end of the PECAM-1 mRNA, were also generated; however, sequence analyzes of several of these products showed three additional transcription initiation sites, all within eight nucleotides of the A immediately 3 'of the nucleotide in SEQ ID NO: l. The location of the transcription start site for this region is consistent with the discoveries of Zehnder et al., Supra, who reported a cDNA clone containing a region not translated in 5 'of 207 bp. Therefore, it seems that the PECAM-1 gene (the part of the PECAM-1 genomic DNA that is transcribe) starts (and therefore, the transcription start site of the PECAM-1 genomic DNA is in) one of the different nucleotides closely separated, similar to the situation found for the genes for vascular cell adhesion molecules. -selectin (Collins et al. (1991) J. Biol. Chem. 266, 2466-2473), and L-selectin (Ord et al. (1990) J. Biol. Chem. 265, 7760-7767). The main characteristics of the PECAM-1 gene were found as follows. The gene is composed of 16 exons separated by introns ranging from 86 to more than 12,000 base pairs in length. Exon 1, which corresponds to the 5 'untranslated region (UT) of the PECAM-1 cDNA, also codes for the mRNA encoding most, but not all, of the signal peptide. Exon 2, which resides in close proximity to exon 1 on the gene, and is only 27 bp in length, encodes the RNA encoding the rest of the signal peptide, and the first three amino acids of the predicted N-terminus of the protein mature Subsequently, there is a direct correlation between the exon / intron organization and the structure of the extracellular domain of the PECAM-1 protein. Similar to other members of the Ig superfamily, each of the six homology domains with Ig (see Figure 2 of the '554 patent) corresponds to its own exon, numbered from 3 to 8, which encodes the mRNA that codes for the homology domain. The transmembrane segment (amino acids 575-593 in Figure 1 of the '554 patent) with its immediate flanking segments, also corresponds to a separate exon, exon 9. In an unexpected manner, we discovered that the cytoplasmic tail of PECAM-1 It is divided into seven distinct segments / regions, each of which corresponds to an exon of the PECAM gene. Accordingly, each of seven exons, exons 10-16, of the PECAM-1 gene, codes for the mRNA encoding a segment of the cytoplasmic tail. Exon 16 codes for the 3 'UT (untranslated region) of transcription of PECAM-1 mRNA. The nucleotide sequence for 871 additional bp of the PECAM-1 genomic DNA which is 3 'for the triplet of the gene encoding the translation stop codon was also determined. In this segment of 871 bp, a primary 5'-AATAAA polyadenylation sequence was not found in consensus. However, two consensus sequences, 5'-GATAAA and 5'-AATACA, were noted after the triplet coding for the stop codon. Alternative splicing (sometimes referred to herein as a "differential" splice) of transcription of the PECAM-1 gene was discovered. Each intron begins with the consensus splice donor sequence "GT", and ends with the consensus splice acceptor sequence "AG". Is Interestingly, all exons, with the exception of exons 10 and 15, end with the first base of the triplet encoding a codon, thus having the classification of "phase 1" exons (Sharp (1981) Cell 23, 643- 646). In particular, five of seven exons that correspond to the transmembrane and cytoplasmic domains (exons 9 - 15), as well as exon 8, are of the phase 1 class, leading to the possibility of alternative splicing events within the framework, which produce isoforms of PECAM-1 that are soluble, due to the lack of a transmembrane domain, or that differ in the sequence of the cytoplasmic tail. In order to examine if these isoforms could be generated inside the cell, the HUVEC mRNA was subjected to amplification with polymerase chain reaction at room temperature, from a region encoded by exons 11-16, using a primer with the sequence of SEQ ID NO: 14, and a primer with the sequence of SEQ ID NO: 15. Two products of the polymerase chain reaction were isolated by agarose gel electrophoresis: a product greater than 260 bp, which corresponds to a full-length segment containing all exons 12-15, part of exon 11, and part of exon 16, and a product less than 200 bp that was hybridized with a full length PECAM-1 cDNA probe. When the minor product of the polymerase chain reaction, derived from the mRNA, was subcloned and sequenced, it was found to be the same as the full-length product, except that exon 14 was missing. Therefore, the mRNA corresponding to this minor product encodes an isoform of PECAM-1α. An "isoform of PECAM-l? X" is the isoform that has the amino acid sequence that results from the translation of the mRNA that is the same as a mRNA that moves to make a full-length PECAM-l, except that the mRNA that is transferred to make the isoform lacks the mRNA segment that corresponds to exon x of the gene for PECAM-1. In a similar way, an "isoform of PECAM-l? X, y, ... z" is the isoform that has the amino acid sequence that results from the translation of the mRNA that is the same as an? RNm that moves to make a full length PECAM-l, except that the mRNA that is transferred to make the isoform lacks the mRNA segments corresponding to the exons x, y, ... and z, of the gene for PECAM-1. By examining the exon organization of the PECAM-1 gene, it has been determined that the soluble form of PECAM-1 identified by Goldberger et al., Supra, is an isoform of PECAM-1 encoded by an mRNA lacking the segment encoded by exon 9 of the genomic DNA (ie, an isoform of PECAM-1? 9). This missing segment in this mRNA, which corresponds to exon 9, includes a subsegment that encodes the transmembrane domain, amino acids 575 - 593 of the PECAM-1, as shown in Figure 1 of the '554 Patent. A cDNA has been discovered that differs from the cDNA for which the sequence is reported in Figure 1 of the '554 patent, substantially only because it lacks the 63 base pair segment corresponding to the bases 2186-2228 shown in that Figure 1 This segment corresponds exactly to exon 13 of the PECAM-1 gene, for which the cDNA sequence is provided in Figure 1 of the '554 patent. Therefore, an isoform of PECAM-1? 13 has been discovered. Because exons 12, 13 and 14 are all phase 1 exons, splicing the mRNA encoding PECAM-1, the segment encoded by exon 13, results in an accurate suppression of 21 amino acids, without changing the framework of reading the mRNA that encodes the rest of the cytoplasmic tail. The amino acid segment corresponding to exon 13 contains 4 of the 12 serine residues found in the cytoplasmic domain of PECAM-1, and thus, when present, it may serve as a site for post-translational phosphorylation and the cytoskeletal association of the PECAM-l. Therefore, the isoform of PECAM-1? 13 must have a differential capacity, in comparison with PECAM-1 itself or other isoforms thereof, to become phosphorylated and associated with the cytoskeleton.

Others have recently isolated and characterized an isoform of murine PECAM-1, a murine PECAM-12, 15, from the development of cardiac endothelium. This isoform has an amino acid sequence that differs from that of the full-length mature molecule, because it is encoded by an RNA that differs from the RNA encoding the full-length molecule, lacking the segments corresponding to exons 12 and 15 of the gene for murine PECAM-1. Therefore, it would be noted that the highly divided exon / intron organization of the PECAM-1 gene in the region corresponding to the cytoplasmic tail is actively used by the cells where the PECAM-1 is made for the generation of PECAM isoforms. -l. Surprisingly, the multi-exon mosaic structure of the PECAM-1 genomic DNA extends to the cytoplasmic region. Unlike the genes for the homologous proteins, ICAM-1 and VCAM-1, which combine the coding regions for the transmembrane and cytoplasmic domains in a single exon, it has been found that the cytoplasmic tail of the PECAM-1 is encoded by an RNA encoded by seven separate small exons, each of which appears to represent a domain with a separate function.

EXAMPLE 2 In this example, a method is provided for removing from a cDNA for a PECAM-1, or an isoform thereof, a previously determined particular segment. In this example, the method is used to eliminate the cDNA segment corresponding to bases 44-100 of exon 9 shown in SEQ ID NO: 4. This cDNA segment codes for the mRNA segment encoding PECAM-1, that encodes the "transmembrane domain". An isoform of PECAM-1 expressed from a cDNA lacking this segment is soluble. The skilled person could easily adapt the method described herein to eliminate, from a cDNA segment encoding an mRNA encoding a full length PECAM-l or an isoform thereof, the segment corresponding to any one or more of the exons 9-15. The information required to adapt the method in this manner is the information provided herein on the sequences of the exons in the genomic DNA of the PECAM-1 (and, therefore, the limits of exons in the cDNAs. from PECAM-1), and information readily available in the art, including that on cDNA sequences for full length PECAM-l (eg, the '554 patent, Stockinger et al., supra; and Zehnder et al., supra; ), the recognition sequences of restriction enzymes and dissociation sites, and the sequences of many vectors of adequate cloning. The technique described herein is based on that described by Kahn (1990) Technique - A Journal of Methods in Cell and Molecular Biology 2, 27-30. The PECAM-1 cDNA, whose sequence is illustrated in Patent Figure '554, was maintained in a vector between two EcoRI sites. Actually, the 5 -GAATT shown in that Figure at the 5 'end of the cDNA, is part of an EcoRI site, and an artifice of the method for the preparation of the cDNA library, from which the cDNA was prepared, and the isolation of the cDNA from the library. The PECAM-1 cDNA, with the sequence shown in Figure 1 of the '554 patent, was converted by site-directed utagenesis into a more preferred cDNA, which conveniently is without the two internal EcoRI sites in the cDNA with the sequence of Figure 1. The sequence of the EcoRI site at positions 684-689 in that Figure was converted to 5'-GAACTC. The sequence of the site at positions 715-720 in that Figure was converted to 5'-GAGTTC. The two changes of bases are silent. The resulting cDNA is referred to herein as PECAM-1"minus EcoRI cDNA of Figure 1". The PECAM-1 minus EcoRI cDNA of Figure 1 was cloned into the EcoRI site of a plasmid vector pGEM-7 (Promega), where it is conveniently maintained for production, modification, and separation with EcoRI, for transfer to other vectors, including expression vectors, or for other purposes. Many other vectors available in the art would be suitable, rather than a pGEM-7 vector, also for these purposes. A 343 base pair segment, bases 1602-1944, of the sequence shown in Figure 1 of the '554 Patent, of the PECAM-1 cDNA minus EcoRI of Figure 1, in vector pGEM-7, was amplified by polymerase chain reaction, using a primer with the sequence of SEQ ID NO: 16, and a primer with the sequence of SEQ ID NO: 17. The product of the amplification was a segment of 361 base pairs, referred to in FIG. present as "Segment A", which consisted of the segment of 343 base pairs at one end, and a segment of 18 base pairs at the other end, with the base sequence 2002-2019, as shown in Figure 1 of the '554 Patent. Segment A was isolated from the reaction mixture. Note that the cDNA segment with the base sequence 1945-2001, as shown in Figure 1 of the '554 patent, corresponds to the segment of exon 9 with the base sequence 44-100 shown in SEQ ID NO: 4 A segment of 481 base pairs, bases 2002 -2482, of the sequence shown in Figure 1 of the '554 Patent, of the PECAM-1 cDNA minus EcoRI of Figure 1 in vector pGEM-7, was amplified by chain reaction of polymerase, using a primer with the sequence of SEQ ID NO: 18, and a primer with the sequence of SEQ ID NO: 19. The product of the amplification was a 497 base pair segment, referred to herein as " Segment B ", which consisted of the segment of 481 base pairs at one end, and at the other end, a segment of 16 base pairs, with the sequence of bases 1929 - 1944, as shown in Figure 1 of Patent '554. Segment B was isolated from the reaction mixture. Note that the sequence of a segment of 34 base pairs at one end of Segment A is the same as that of a segment of 34 base pairs at one end of Segment B. As a result, each Segment A chain can prime a primer extension reaction on one of the Segment B chains as a template, and each Segment B chain can prime a primer extension reaction on one of the Segment A chains as a template. Then a polymerase chain reaction was performed with the combination of Segment A, Segment B, a primer with the sequence of SEQ ID NO: 20, and a primer with the sequence of SEQ ID NO: 21. The major product of this reaction was a 533 base pair segment, referred to herein as "Segment C", with a subsegment at one end with the sequence of the bases 1754-1944 as shown in Figure 1 of the '554 patent, and one subsegment in the other end with the sequence of the bases 2002 - 2343 as shown in that Figure 1. Segment C has a Nhel site (5'-GCTAGC, with dissociation between G and C at the 5 'end), as the subsegment corresponding to the bases 1825-1830 shown in Figure 1 of the '554 patent, and a BglII site (5'-AGATCT, with dissociation between A and G at the 5' end) as the subsegment corresponding to bases 2247-252 shown in Figure 1 of the '554 Patent. Segment C was cut with Nhel and BglII, and the fragment approximately 369 bp was isolated. The PECAM-1 minus EcoRI cDNA from full-length Figure 1 was also cut in the vector pGEM-7, with Nhel and BglII, and the larger fragment was isolated. Note that the vector pGEM-7 itself does not have sites for Nhel and BglII. The 369 base pair fragment of Segment C was then ligated into the larger fragment from the Nhel / BglII dissociation of vector pGEM-7 with the PECAM-1 cDNA minus EcoRI of Figure 1. The resulting vector had CDNA for a soluble PECAM-1 isoform, which had the entire amino acid sequence shown in Figure 1 of the '554 patent, except for the amino acids of the "transmembrane domain", amino acids 575-593, as shown in that Figure , or alternatively, amino acids 9-27 shown in SEQ ID NO: 4.

SEQUENCE LIST i _ 2 ID NO: l TCTTTTTGGTT TTGCTATTGC TTAAGCTAGC CTACGCCAAG GGTGCTCTTT GCCCCCTACT 60 TCCTCTGCTA TTCTCGCCTC AGTTCCGCTG CATTCCAAGC TCAGCCTGCC CCAGCAGCAG 120 GTCTCTTTGA CAAACCTGCA ATTTTGGCGA AAAGTCAGCC CAAGAAAGGC AGGGGGCCCC '180 GACTTATGCT GTGTGGCAAA AGCCCTCTTT GATGGGGCAA GGGTAGGACT GGAAAAGCAG 240 AGAGATCTTT CTGGATGTCC TGGGAGAGCA GCCCTTTGGG TGGTGGGTGG AGGCTGGAGG 300 CAGGGAGGAA TCCCCTCACA GTGAGAAGGG CCCCCAAACC CAGGCGAGAC AGAGGGAGGG 360 TCAAGAACGC CAAGGCAAAT GTCACTTGTG CCTTGTTTTT TCCCTAAAGA AACTAAACAA 420 AGCGGCCGCG TTCGGTGGCC CCTCAGGAAG GCCGGTCATT TCCTGAGGAG ATATCAGGCC 480 AGCCCAGGCC CC 492 SEQ ID NO: 2 CCTGGGAGAG CAGCCCTTTG GGTGGTGGGT GGAGGCTGGA GGCAGGGAGG AATCCCCTCA 60 CAGTGAGAAG GGCCCCCAAA CCCAGGCGAG ACAGAGGGAG GGTCAAGAAC GCCAAGGCAA 120 ATGTCACTTG TGCCTTGTTT TTTCCCTAAA GAAACTAAAC AAAGCGGCCG CGTTCGGTGG 180 CCCCTCAGGA AGGCCGGTCA TTTCCTGAGG AGATATCAGG CCAGCCCAGG CCCC 234 SEQ ID NO: 3 CC CCG GTG GAT GAG GTC CAG ATT TCT ATC CTG TCA AGT AAG GTG GTG 47 Wing Pro Val Asp Glu Val Gln lie Ser Xle Leu Ser Ser yß Val Val 5 10 15 CAG TCT GGA GAG GAC ATT GTG CTG CA TGT GCT GTG AAT GAA GGA TCT 95 Gly Ser Gly Glu Aßp lie Val Leu Gln Cyß Wing Val Asn Glu Gly Ser 20 25 30 GGT CCC ATC ACC TAT AAG TTT TAC AGA GAA AAA GAG GGC AAA CCC TTC 143 Gly Pro He Thr Tyr Lyß Phß Tyr Arg Glu Lyß Glu Gly Lyß Pro Phß 35 40 45 TAT CAA ATG ACC TCA AAT GCC ACC CAG GCA TTT TGG ACC AAG CAG AAG 191 Tyr G n Met Thr Ser Asn Wing Thr Gln Wing Phe Trp Thr Lyß Gln Lyß 50 55 60 GCT AAC AAG GAA CAG GAG GAC TAC TAC TGC ACA GCC TTC AAC AGA 239 Wing Asn Lyß Glu Gln Clu Gly Glu Tyr Tyr Cyß Thr Wing Phe Asn Arg 65 70 75 80 GCC AAC CAC GCC TCC AGT GTC CCCA ATA CTG ACA GTC AGA 287 Ala Asn His Ala Ser Ser Val Pro Arg Ser Lys He Leu Thr Val Arg 85 90 95 G GTGAGTCAGG GTCTCCATAG_308_^ TGTTTTTTTT TTTTTTTCAG TC ATT CTT GCC CCA TGG AAG AAA GGA CTT ATT 52 Val He Leu Pro Wing Trp Lyß Lyß Gly Leu He 5 10 GCA GTG GTT ATC ATC GGA GTG ATC ATT GCT CTC TTG ATC ATT GCG GCC 100 Wing Val Val He He Gly Val He He Wing Leu Leu He Wing Wing 15 20 25 AAA TGT TAT TTT CTG AGG AAA GCC AAG G GTGAGCATAG TTCTTTCCTT 148 Lys Cyß Tyr Phe Leu Arg Lys Ala Lyß 30 35 SEQ ID NO: 5 TTCGTTTTCT GTTTTTAAAG CC AAG CAG ATG CCA GTG GAA ATG TCC AG 48 Ala Lys Gln Het Pro Val Glu Met Ser Arg 5 10 GTGAGTGTAT TTGTAAGAAG 68 SEQ ID NO: 6 TTTTATATTT CATTTTAAAG G CCA CCA GTA CCA CTT CTG AAC TCC AAC AAC 51 Pro Ala Val Pro Leu Leu Aßn Ser Aßn Asn 5 10 GAG AAA ATG TCA GAT CCC AAT ATG CAAC GCT AAC AGT CAT TAC G 94 Glu Lyß Met Ser Asp Pro Aßn Met Glu Ala Aßn Ser Hiß Tyr 15 20 GTAAAGTCAT GTTCTCCTGC 114 SEQ ID NO: 7 AATTGTTATT TTTCAACTAG GT CAC AAT GAC GAT GTC AGA AAC CAT GCA ATG 52 Gly His Asn? ßp Aßp Val Arg Aßn Hiß Ala Met 5 10 AAA CCA ATA AAT GAT AAT AAA G OTAATTATCT AATTACATGT 94 SEQ ID NO: 8 TCTGTGGTTT CTTTAGGCAG AG CCT CTG AAC TCA GAC GTG CAG TAC ACG GAA 52 Glu Pro Leu Aßn Ser Aßp Val Gln Tyr Thr Glu 5 10 GTT CAÁ GTG TCC TCA GCT GAG TCT CAC'AAA G GTAAGTGCCA CTCGAGTGAG 103 Val Gln Val Ser Ser Ala Glu Ser Hiß Lyß 15, 20 SEQ ID NO: ^ ÍGCCTGGTC CTTTTTCCAG AT CTA GGA AAG AAG GAC ACA GAG ACA GTG TAC 52 Asp Leu Gly Lys Lyß Asp Thr Glu Thr Val Tyr 5 10 AGT GAA GTC CGG AAA GCT GTC CCT G GTGAGTGAGG GTCTCCAGTG 97 Ser Glu Val Arg Lyß Ala Val Pro 15 SEQ ID NO: 10 TGTCATCCTT TGTTTTGTAG AT GCC GTG GAA AGC AGA TAC TCT GTAAGTACAC 53 4 Aßp Ala Val Glu Ser Arg Tyr Ser 5 ATTTCATATA 63 SEQ ID NO: 11 CTTGTTTCTT GTCGCTACAG AGA ACG GAA GGC TCC CTT GAT GGA ACT TAG_50_Arg Thr Glu Gly Ser Leu Asp Gly Thr 5 ACAGCAAGGC CAGATGCACA TCCCTGGAAG GACATCCATG TTCCGACAAG AACAGATGAT 110 CCCTGTATTT CAAGACCTCT GTGCACTTAT TTATGAACCT GCCCTGCTCC CACAGAACAC 170 AGCAATTCCT CAGGCTAAGC TGCCGGTTCT TAAATCCATC CTGCTAAGTT AATGTTGGGT 230 AGAAAGAGAT ACAGACGGCC TGTTGAATTT CCCACATACC CTCCTTCCAC CAAGTTGGAA 290 CATCCTTGGA AATTGGGAAG AGCACAAGAG GAGATCCAGG GCAAGGCCAT TGGGATATTC 350 TGAAACTTGA ATATTTTGTT TTGTGCAGAG ATAAAGACCT TTTCCATGCA CCCTCATACA 410 CAGAAACCAA TTTTCTTTTT TATACTCAAT CATTTCTAGC GCATGGCCTG GTTAGAGGCT 470"GGTTTTTTTCT CTTTTCCTTT GGTCCTTCAA AGGCTTGTAG TTTTGGGTAG TCCTTGTTCT 530 TTGGAAATAC ACAGTGCTGA CCAGACAGCC TCCCCCTCTC CCCTCTATGA CCTCGCCCTC 590 CACAAATGGG AAAACCAGAC TACTTGGGAG CACCGCCTGT GAAATACCAA CCTGAAGACA 650 CGGTTCATTC AGGCAACGCA CAAAACAGAA AATGAAGGTG GAACAAGCAC ATATGTTCTT 710 CAACTGTTTT TGTCTACACT CTTTCTCTTT TCCTCTACAT GCTGAAGGCT GAAAGACAGG 770 AAAGATCGTG CCATCAGCAA ATATTATTCT TAATTGAAAA CTTGAAATCT GTATGTTTCT 830 TACTAATTTT TAAAAATGTA TTCCTTGCCA GGCCAGGCAA GGTCGTCACG CCTGTAATCC 890 CAGCACTTCA GGAGGCTGAG GTGGGCGGAT C '921 SEQ ID NO: 12 CCTGAGAGTG AAGACTGCAG GCACAGTTAG TTCTGCCTTC GG 42 SEQ ID NO: 13 TGCTGTGTTC TGTGGGAG 18 ß £ Q ID NO: 14 «? ACGAGAAA ATGTCAGA 18 S? Q ID NO: 15 GGAGCCTTCC GTTCTAGAGT .20 SEQ ID NO: 16 GTTAAGTGAG GTTCJTGAGGG 20 SEQ ID NO: 17 ACGGGGTACC TTCTTTTTTA CAATAAAAGA CTCC 34 SEQ ID NO: 18 TGCCCCATGG AAGAAAAAAT GTTATTTTCT GAGG 34 SEQ ID NO: 19 TGCTGTGTTC TGTGGGAG 18 SEQ ID NO: 20 GAGAAAAAGA GGGCAAACCC 20 SEQ ID NO: 21 GGAGCCTTCC GTTCTAGAGT 20

Claims (20)

1. An isolated DNA molecule comprising the nucleotide sequence of SEQ ID NO: 2, wherein said DNA molecule is the promoter of the platelet endothelial cell adhesion molecule-1 (PECAM-1).

2. An isolated DNA molecule comprising a nucleotide sequence that is substantially the same as the nucleotide sequence of SEQ ID NO: 2, wherein the DNA molecule is capable of stimulating transcription within a mammalian cell of the vasculature, at an index that is at least 10 percent of the transcription index stimulated by a DNA molecule having the nucleotide sequence of SEQ ID NO: 2 under comparable conditions.

3. The isolated DNA molecule of claim 1, wherein said DNA molecule consists of a nucleotide sequence selected from the group consisting of SEQ ID NO: l and SEQ ID NO: 2.

4. A vector of expression comprising the engine of claim 1.

5. The expression vector of claim 4, which further comprises a second DNA molecule encoding either a polypeptide or an anti-sense RNA, wherein the promoter is it links operably with the second DNA molecule.

6. The expression vector of claim 5, wherein the second DNA molecule encodes the PECAM-1 polypeptide.

7. The expression vector of claim 5, wherein the promoter consists of a nucleotide sequence selected from the group consisting of SEQ ID NO: 2 and SEQ ID NO: 2.

8. An isolated DNA molecule encoding a variant of an isoform of the polypeptide of platelet endothelial cell adhesion molecule-1 (PECAM-1), wherein this isoform is selected from the group consisting of form 1, form 2, form 3, and 4, wherein the isoform of PECAM-1 variant lacks at least one amino acid sequence selected from the group consisting of: (a) the amino acid sequence encoded by SEQ ID NO: 4, (b) the sequence of amino acids encoded by nucleotides 44-100 of SEQ ID NO: 4, (c) the amino acid sequence encoded by SEQ ID NO: 5, (d) the amino acid sequence encoded by SEQ ID NO: 6, (e) ) the amino acid sequence encoded by SEQ ID NO: 7, (f) the amino acid sequence encoded by SEQ ID NO: 8, (g) the amino acid sequence encoded by SEQ ID NO: 9, and (h) the amino acid sequence encoded by SEQ ID NO: 10. 9 The DNA molecule of claim 8, wherein the isoform of PECAM-1 variant is selected from the group consisting of: (a) a polypeptide lacking the amino acid sequence encoded by SEQ ID NO: 8, (b) a polypeptide lacking (i) the amino acid sequence encoded by SEQ ID NO: 8, and (ii) the amino acid sequence encoded by SEQ ID NO: 4, (c) a polypeptide that lacks ( i) the amino acid sequence encoded by SEQ ID NO: 8, and (ii) the amino acid sequence encoded by nucleotides 44-100 of SEQ ID NO: 4, (d) a polypeptide lacking the amino acid sequence encoded by SEQ ID NO: 9, (e) a polypeptide lacking (i) the amino acid sequence encoded by SEQ ID NO: 9, and (ii) ) the amino acid sequence encoded by SEQ ID NO: 4, and (f) a polypeptide lacking (i) the amino acid sequence encoded by SEQ ID NO: 9, and (ii) the amino acid sequence encoded by nucleotides 44-100 of SEQ ID NO: 4. 10. The DNA molecule of claim 8, wherein the isoform of PECAM-1 variant is a polypeptide lacking the encoded amino acid sequence. SEQ ID NO: 4. 11. An expression vector comprising the DNA molecule of claim 8. 12. A method for using the expression vector of claim 11, for preparing a PECAM-1 isoform polypeptide. variant, this method comprising the step of culturing host cells comprising this expression vector. 13. The method of claim 12, wherein the host cells are mammalian cells. 14. An expression vector comprising the DNA molecule of claim

9. 15. A method for using the expression vector of claim 14, for preparing a variant PECAM-1 isoform polypeptide, this method comprising the step of of culturing host cells comprising this expression vector. 16. The method of claim 15, wherein the host cells are mammalian cells. 17. The method of claim 15, wherein theHost cells are bacterial cells or yeast cells. 18. A substantially isolated variant of an isoform of a platelet endothelial cell adhesion molecule-1 polypeptide (PECAM-1), wherein this isoform is selected from the group consisting of form 1, form 2, form 3 and form 4, wherein the isoform of PECAM-1 variant lacks at least one amino acid sequence selected from the group consisting of: (a) the amino acid sequence encoded by SEQ ID NO: 4, (b) the amino acid sequence encoded by nucleotides 44-100 of SEQ ID NO: 4, (c) the amino acid sequence encoded by SEQ ID NO: 5, (d) the amino acid sequence encoded by SEQ ID NO: 6, (e) the amino acid sequence encoded by SEQ ID NO: 7, (f) the amino acid sequence encoded by SEQ ID NO: 8, (g) the amino acid sequence encoded by SEQ ID NO: 9, (g) ) the amino acid sequence encoded by SEQ ID NO:

10. 19. The isoform of PECAM-1 variant of the claim 18, wherein the isoform of PECAM-1 variant is selected from the group consisting of: (a) a polypeptide lacking the amino acid sequence encoded by SEQ ID NO: 8, (b) a polypeptide that is lacking of (i) the amino acid sequence encoded by SEQ ID NO: 8, and (ii) the amino acid sequence encoded by SEQ ID NO: 4, (c) a polypeptide lacking (i) the encoded amino acid sequence by SEQ ID NO: 8, and (ii) the amino acid sequence encoded by lbs nucleotides 44-100 of SEQ ID NO: 4, (d) a polypeptide lacking the amino acid sequence encoded by SEQ ID NO: 9, (e) a polypeptide lacking (i) the amino acid sequence encoded by SEQ ID NO: 9, and (ii) the amino acid sequence encoded by SEQ ID NO: 4, and (f) a polypeptide that lacks (i) the amino acid sequence encoded by SEQ ID NO: 9, and (ii) the amino acid sequence encoded by the nucleotides. two 44-100 of SEQ ID NO: 4. 20. The DNA molecule of claim 18, wherein the isoform of PECAM-1 variant is a polypeptide lacking the amino acid sequence encoded by SEQ ID NO: 4. .