Mutations in TNFRSF11A, affecting the signal peptide of RANK, cause familial expansile osteolysis.

Familial expansile osteolysis (FEO, MIM 174810) is a rare, autosomal dominant bone disorder characterized by focal areas of increased bone remodelling. The osteolytic lesions, which develop usually in the long bones during early adulthood, show increased osteoblast and osteoclast activity. Our previous linkage studies mapped the gene responsible for FEO to an interval of less than 5 cM between D18S64 and D18S51 on chromosome 18q21.2-21.3 in a large Northern Irish family. The gene encoding receptor activator of nuclear factor-kappa B (RANK; ref. 5), TNFRSF11A, maps to this region. RANK is essential in osteoclast formation. We identified two heterozygous insertion mutations in exon 1 of TNFRSF11A in affected members of four families with FEO or familial Paget disease of bone (PDB). One was a duplication of 18 bases and the other a duplication of 27 bases, both of which affected the signal peptide region of the RANK molecule. Expression of recombinant forms of the mutant RANK proteins revealed perturbations in expression levels and lack of normal cleavage of the signal peptide. Both mutations caused an increase in RANK-mediated nuclear factor-kappaB (NF-kappaB) signalling in vitro, consistent with the presence of an activating mutation.

Cloning and expression of the gene for the melanoma-associated ME20 antigen.

Human melanoma cells, but not tumor cells of other histological origin, express a unique membrane-associated glycoprotein, designated ME20-M, and secrete a soluble glycoprotein, designated ME20-S, defined by monoclonal antibody ME20. Here we report the isolation and characterization of a cDNA clone that when transfected into COS cells directs the expression of ME20-M and ME20-S. This cDNA contains an open reading frame which encodes a 661-amino-acid-long precursor that contains a 23-amino-acid signal peptide and a 26-amino-acid transmembrane domain, separated by a hydrophilic region containing 5 potential Asn-linked and 14 predicted Pro-associated, Thr-linked glycosylation sites. The transmembrane domain is followed by a carboxy-terminal 45-amino-acid putative intracellular domain rich in Ser residues. Analysis of ME20-M by amino acid sequencing identified the proteolytic processing site. Signal peptide cleavage occurs at the Thr-24-Lys-25 peptide bond of the precursor and results in the 637-amino-acid ME20-M with a calculated molecular weight of 67,782. ME20-M is derived from a single 3.3- to 3.4-kb mRNA transcript that is expressed at varying levels in melanoma cell lines, correlating with immunofluorescence determination of protein expression. The amino acid sequence of the ME20 antigen deduced from the cDNA differs from the human neonatal melanocyte-specific Pmel 17 gene product by a single amino acid substitution and deletion of 7 amino acid residues, and it is 80% homologous with the bovine retinal pigment RPE1 cDNA.(ABSTRACT TRUNCATED AT 250 WORDS)

Differential effects of cartilage-derived growth factor stimulation of collagen secretion by bovine aortic and microvascular endothelial cells.

Cartilage-derived growth factor (CDGF), a protein closely related to basic fibroblast growth factor, is known to have both mitogenic and chemokinetic properties in microvascular endothelial cells (MVEC). Because of the angiogenic properties of CDGF and its rate in accelerating wound repair, the capacity of this factor to stimulate both proliferation and matrix synthesis was compared in distinct populations of vascular endothelial cells: MVEC from bovine adrenal cortex and macrovascular endothelial cells from the bovine aorta (BAEC). No significant differences in the responses to mitogenic stimulation using CDGF (5-100 units/ml) were observed using MVEC and BAEC. Only rapidly dividing MVEC, however, showed significant increases in collagen secretion in the presence of CDGF. The differential responsiveness of these two cell populations to a defined growth factor underscores the phenotypic diversity of endothelium.

Cloning and expression of the tumor-associated antigen L6.

The L6 cell surface antigen, which is highly expressed on lung, breast, colon, and ovarian carcinomas, has attracted attention as a therapeutic target for murine monoclonal antibodies and their humanized counterparts. Its molecular nature has, however, remained elusive. Here we describe the expression cloning of a cDNA encoding the L6 antigen. COS cells transfected with this cDNA direct the expression of an approximately 24-kDa surface protein that reacts with the two anti-L6 monoclonal antibodies available. The predicted L6 peptide sequence is 202 amino acids long and contains three predicted NH2-terminal hydrophobic transmembrane regions, which are followed by a hydrophilic region containing two potential N-linked glycosylation sites and a COOH-terminal hydrophobic transmembrane region. The L6 antigen is related to a number of cell surface proteins with similar predicted membrane topology that have been implicated in cell growth. Two other members of this family of proteins, CD63 (ME491) and CO-029, are also highly expressed on tumor cells. The present findings should make it possible to further study the role of the L6-defined antigen in normal and neoplastic cells and to construct animal models for development of improved agents for active and passive cancer immunotherapy.

Molecular genetic basis of the histo-blood group ABO system.

The histo-blood group ABO, the major human alloantigen system, involves three carbohydrate antigens (ABH). A, B and AB individuals express glycosyltransferase activities converting the H antigen into A or B antigens, whereas O(H) individuals lack such activity. Here we present a molecular basis for the ABO genotypes. The A and B genes differ in a few single-base substitutions, changing four amino-acid residues that may cause differences in A and B transferase specificity. A critical single-base deletion was found in the O gene, which results in an entirely different, inactive protein incapable of modifying the H antigen.

Cloning and characterization of DNA complementary to human UDP-GalNAc: Fuc alpha 1----2Gal alpha 1----3GalNAc transferase (histo-blood group A transferase) mRNA.

Based on the partial amino acid sequence, the cDNA encoding UDP-GalNAc:Fuc alpha 1----2Gal alpha 1----3GalNAc transferase, the specific primary gene product of histo-blood group A gene (A transferase), was cloned and sequenced. Poly(A)+ RNA from human stomach cancer cell line MKN45, expressing high levels of A antigen, was used for construction of a lambda gt10 cDNA library. Degenerate synthetic oligodeoxynucleotides were used for polymerase chain reactions to detect the presence of the sequence of interest in cDNA (presence test) and to identify the correct clones (identification test) after screening the library with a radiolabeled polymerase chain reaction amplified fragment. Nucleotide sequence analysis revealed a coding region of 1062 base pairs encoding a protein of 41 kDa. Hydrophobicity plot analysis shows the existence of three domains: N-terminal short stretch, transmembranous hydrophobic region, and a long C-terminal domain (a feature common to all glycosyltransferases cloned so far). Southern hybridization analysis has shown that this DNA does not represent a multigene family. No restriction fragment length polymorphism was found to correlate with ABO blood group type. Bands were detected in Northern hybridization of mRNAs from cell lines expressing A, B, AB, or H antigens. These results suggest that sequences of ABO genes are essentially very similar (with minimal differences), and the inability of the O gene to encode A or B transferases is probably due to structural differences rather than A or B transferase expression failure.

Chromosomal localization of three human genes coding for A15, L6, and S5.7 (TAPA1): all members of the transmembrane 4 superfamily of proteins.

The A15, L6, and S5.7(TAPA1) proteins are members of the transmembrane 4 superfamily (TM4SF). The A15 is expressed in immature human T cells and in the human brain. The MXS1(CCG-B7) gene which codes for A15 contains triplet nucleotide repeats which have been associated with neuropsychiatric diseases such as Huntington's chorea, fragile X syndrome, and myotonic dystrophy. The L6 antigen is mainly expressed in lung, breast, colon, ovarian carcinomas, and healthy epithelial tissue in humans. The S5.7(TAPA1) antigen is expressed in most human cell lines and is shown to be associated with B-cell surface molecules CD19 and Leu-13. In this study we have used interspecies specific somatic cell hybrids and human-specific cDNA probes to localize the A15 (MXS1), L6 (M3S1), and TAPA1 genes to Xq11, 3q21-25, and 11p15.5, respectively.

Cloning of the murine counterpart of the tumor-associated antigen H-L6: epitope mapping of the human and murine L6 antigens.

The murine monoclonal antibody (mAb) L6 was raised against human lung carcinoma cells and found to recognize an antigen which is highly expressed on lung, breast, colon, and ovarian carcinomas. Promising results in phase 1 clinical studies with this antibody or its chimerized counterpart suggest the antigen recognized by mAb L6 (H-L6) is an attractive target for monoclonal antibody-based cancer therapy. Further development of L6 as an anti-tumor-targeting agent would benefit from the development of a murine model. However, initial attempts to develop such a model were hampered by our inability to generate antibodies against the murine homologue of the L6 antigen, M-L6. Here we describe the preparation of the mAb 12A8, which was raised against murine thymic epithelial cells, the tissue distribution of the murine antigen recognized by 12A8, the cloning of a cDNA encoding the 12A8 target antigen, and the demonstration that this antigen is M-L6. Using H-L6/M-L6 chimeric proteins, we show that the region of the M-L6 protein recognized by mAb 12A8 corresponds to the region of H-L6 recognized by mAb L6. There are five amino acid differences in the regions of the H-L6 and M-L6 proteins recognized by L6 and 12A8, respectively. We further mapped the protein epitope recognized by L6 by individually exchanging each of these residues in H-L6 with the corresponding residue found in M-L6. Substitution of the single H-L6 residue Leu122 with Ser resulted in the H-L6 mutant HL6-L122S which failed to bind L6. The HL6-L122S mutant also failed to bind 12A8.(ABSTRACT TRUNCATED AT 250 WORDS)

Analysis of gp39/CD40 interactions using molecular models and site-directed mutagenesis.

The interaction between gp39 (CD40L, TRAP, T-BAM) on activated T cells and mast cells and CD40 on antigen-presenting cells modulates immune responses. Gp39 and CD40 are homologous to tumor necrosis factor (TNF) and its receptor (TNFR), respectively. The TNF-beta/TNFR interaction has been analyzed on the basis of mutagenesis experiments and crystal structures. Using the interaction of TNF-beta/TNFR as a guide, we previously reported a site-directed mutagenesis study in which we identified residues in gp39 (K143, Y145) and CD40 (Y82, D84, N86) involved in gp39/CD40 interactions. Here we describe the use of the TNF-beta/TNFR complex crystal structure as a template to prepare molecular models of gp39, CD40, and their approximate interaction. The application of these models has allowed us to extend our mutagenesis analysis of gp39/CD40 interactions. These experiments have led to the identification of additional gp39 (Y146, R203, Q220) and CD40 (E74, E117) residues that contribute to the gp39/CD40 interaction. We also further explored the importance of gp39 residue Y145 and CD40 residue Y82 for the gp39/CD40 interaction by conservatively replacing these residues with Phe. The results of these studies have enabled us to approximately outline the binding sites in gp39 and CD40. It appears that the gp39/CD40 interaction is centered on at least two clusters of residues and involves residues of two adjacent gp39 monomers. The molecular regions involved in the gp39/CD40 interaction essentially correspond to those in the homologous TNF-beta/TNFR system.

Identification of residues on CD40 and its ligand which are critical for the receptor-ligand interaction.

Interactions between gp39 (CD40L, TRAP, T-BAM) on activated T cells and CD40 on antigen-presenting cells play an important role in regulating antibody production by B cells, cytokine production by monocytes, and other immune responses which require T cell "help". Using structure-based sequence alignments, a molecular model of gp39, site-directed mutagenesis, and receptor-ligand binding assays, we have identified CD40 and gp39 surface residues which are important for receptor-ligand binding. Binding studies with CD40 or gp39 proteins containing single and double amino acid substitutions showed that CD40 residues Y82, D84, and N86 are involved in gp39 binding, while gp39 residues K143 and Y145 are important for CD40 binding. Analysis of the location of amino acid substitutions in the naturally occurring gp39 mutants expressed by the X-linked hyper-IgM (X-HIM) patients studied to date indicated the E129/G substitution found in the S128/R-E129/G double mutant affects a solvent-accessible residue which might participate in CD40/gp39 binding. Binding studies with E129/G and E129/A gp39 point mutants showed that this residue does not contribute directly to CD40/gp39 binding but that its substitution with a glycine disrupts the gp39 structure. Comparison of the gp39 and CD40 residues involved in receptor-ligand contacts with those previously identified as playing an important role in TNF-beta/TNFR binding suggests that some of the identified residues from contacts similar to those found in the TNF-beta/TNFR while others are unique to the CD40-gp39 interaction.

Interleukin-6 signal transducer gp130 mediates oncostatin M signaling.

Oncostatin M (OM) is a multifunctional cytokine that is structurally and functionally related to interleukin 6 (IL-6) and leukemia inhibitory factor (LIF). The specific receptor for OM has been demonstrated (by chemical cross-linking) to be a 150-kDa protein in a number of cell lines. The IL-6 signal transducer, gp130, is also an affinity converter for the LIF receptor. It does not bind to either IL-6 or LIF, but associates with the alpha subunits of the receptors and transduces the signals. We examined the possible involvement of gp130 in OM binding and signaling. We demonstrate that: (a) anti-gp130 monoclonal antibodies (mAbs) block the inhibitory effect of OM on A375 cell growth, (b) the binding and cross-linking of 125I-OM to H2981 cells are completely abolished by anti-gp130 mAbs, (c) the cross-linked OM-receptor complex is immunoprecipitated by anti-gp130 mAbs, and (d) COS-7 cells transfected with the full-length cDNA encoding gp130 exhibit increased OM binding and cross-linking, which are also blocked by anti-gp130 mAbs. Therefore, we conclude that the 150-kDa OM binding protein previously characterized in a variety of cell lines is gp130. OM is the natural ligand for gp130 and gp130 mediates the biological responses of OM.

Chimeric L6 anti-tumor antibody. Genomic construction, expression, and characterization of the antigen binding site.

We report the cloning of the genomic variable region genes of the human carcinoma reactive murine monoclonal antibody L6 and their genetic linkage to human constant region exons to encode a human IgG1/kappa chimeric antibody. The chimeric protein was produced at levels greater than 20 micrograms/ml (enabling the initiation of clinical trials) and was found to have binding properties identical with that of the murine parent. The nucleic acid sequence of the variable regions was determined and found to be different than that previously reported (1). The deduced amino acid sequence was then used to generate a structural homology based three-dimensional model of the antibody binding site, which was found to share features with antibodies known to interact with a protein surface, but distinct from those that bind to carbohydrate epitopes. Biochemical analysis of binding between antibody and the in vitro-translated product of a cDNA clone that confers L6 immunoreactivity demonstrates that the antibody recognizes a protein epitope encoded by this transcript which requires the presence of membranes, but is unaffected by the removal of carbohydrate side chains.

Membrane topology of the L6 antigen and identification of the protein epitope recognized by the L6 monoclonal antibody.

The murine monoclonal antibody (mAb) L6 recognizes an integral membrane glycoprotein that is highly expressed on lung, breast, colon, and ovarian carcinomas and is referred to as the L6 antigen. This antigen is an attractive target for therapeutic intervention due to its high level expression on malignant cells. We have previously reported the isolation of a cDNA encoding the human L6 antigen (H-L6). Here, we report the isolation of a cDNA clone encoding the murine L6 antigen (M-L6). This cDNA contains one long open reading frame, which encodes a 220-amino acid polypeptide that is 78% homologous to H-L6. This protein contains short NH2- and COOH-terminal hydrophilic domains and four hydrophobic regions, each long enough to span the plasma membrane. Each of these hydrophobic domains is separated by a hydrophilic domain, the longest of which contains one possible N-linked glycosylation site and is located between the third and fourth hydrophobic domains. We have previously demonstrated that the murine L6 mAb recognizes a protein epitope expressed on human tumor-derived cell lines. Now, using chimeric cDNA constructs encoding human-murine L6 antigen hybrids in conjunction with monoclonal antibody binding experiments, we show that the 42-residue hydrophilic domain of the L6 antigen, located between the third and fourth hydrophobic domains, is outside the cell and that residues in the NH2-terminal region of this domain are critical for the binding of the murine L6 mAb to H-L6.