De novo expression of intercellular-adhesion molecule 1 in melanoma correlates with increased risk of metastasis.

The 89-kDa cell surface glycoprotein, P3.58, is detectable on advanced human melanomas in situ but not on benign melanocytes or early melanomas. cDNA cloning of P3.58 from melanoma cells was accomplished by screening a lambda zap expression vector library with monoclonal antibodies produced against the denatured antigen. Nucleotide sequencing of the clones revealed that P3.58 is identical to the intercellular-adhesion molecule 1. No qualitative differences in P3.58 mRNA species could be seen between melanoma cells and hematopoietic cells and no differences in gene organization were observed between peripheral blood leukocytes and melanoma cells. Inspection of the deduced amino acid sequence of P3.58 indicated the presence of the consensus sequence characteristic for complement-binding proteins. The acquisition of this cell-adhesion molecule during the process of tumor progression is speculated to contribute to the development of metastasis in melanoma.

Structural and functional relationships among receptors and regulators of the complement system.

The classical and alternative pathway of complement activation are regulated by a series of fluid phase and cell-bound factors, some of which at the same time serve as receptors for fragments of C3 and C4. These molecules are factor H, CR1 (C3b/C4b receptor), CR2 (C3d/EBV receptor), C4BP (C4b binding protein), DAF (decay accelerating factor), MCP (membrane cofactor protein; earlier designated p45/70), CR3 (iC3b receptor or Mac-1) and CR4 (protein 150/95). Due to structural, genetic and functional features these factors are members of one or several newly recognized large families of proteins: (1) molecules with 60 amino acids long repeats (H, CR1, CR2, C4BP, DAF); (2) proteins with 1,2-diacylglycerol membrane anchoring (DAF); (3) proteins with a heterodimer structure and preference for ligands containing the tripeptide arginine-glycine-asparagine (CR3, CR4). Recognizing the above mentioned regulators and receptors of the complement system as belonging to these protein families opens new perspectives for further genetic and functional research of mutual interest to complement and noncomplement scientists.

Human complement factor H: isolation of cDNA clones and partial cDNA sequence of the 38-kDa tryptic fragment containing the binding site for C3b.

We isolated cDNA clones coding for the functionally important tryptic N-terminal 38-kDa fragment of human complement control protein factor H using polyclonal and monoclonal antibodies to screen a human liver cDNA library cloned in a bacterial expression vector, PEX-1. By testing the reactivity of antibodies specific for the recombinant proteins produced by individual clones with proteolytic fragments of serum H the exact position of these cDNA clones within H was mapped. One clone, H-19, coding for the 38-kDa fragment of H was sequenced and found to code for 289 amino acids derived from the 38-kDa N-terminal fragment as well as for the first 108 amino acids belonging to the complementary 142-kDa tryptic fragment. The derived protein sequence could be arranged in 6 highly homologous repeats of about 60 amino acids each, the homology between the repeats being determined by the characteristic position of cysteine, proline, glycine, tyrosine and tryptophane residues. The region coding for the epitope recognized by one of our monoclonal antibodies was localized by subcloning restriction fragments of H-19 into the expression plasmid and testing for the expression of this epitope.

Use of a high-efficiency expression vector to isolate cDNA clones for factor H and map their positions within the molecule.

A human liver cDNA library, cloned in a novel high-efficiency bacterial expression vector (PEX), was screened with an affinity-purified antibody to human factor H. Four distinct cDNA clones, H-2, H-40, H-46 and H-49, were identified. Of these, H-2 also reacted with two monoclonal antibodies to H, MAH-4 and OX-24, which were previously shown to recognize the 38,000 N-terminal tryptic fragment of H, carrying the binding site for C3b. By using polyclonal antibodies specific for the domains in H coded for by these cDNA-clones, it could be established that H-2 codes only for the 38,000 N-terminal tryptic fragment of H, whereas H-40, H-46 and H-49 are derived from the 142,000 C-terminal fragment of H. By subcloning H-2 the epitope for OX-24 could be localized as being coded near the central Sma-site of H-2.