The membrane antigen B7/BB1 (refs 1, 2) is expressed on activated B cells, macrophages and dendritic cells, and binds to a counter-receptor, CD28, expressed on T lymphocytes and thymocytes. Interaction between CD28 and B7 results in potent costimulation of T-cell activation initiated through the CD3/T-cell receptor complex. Discrepancies between results with anti-CD28 and anti-B7 antibodies have suggested the existence of a second ligand for CD28 and CTLA-4 (refs 3, 6-8). We have generated a monoclonal antibody, IT2, that reacts with a 70K glycoprotein (B70). B70 complementary DNA was cloned from a B-lymphoblastoid cell line library and encodes a new protein of the immunoglobulin superfamily with limited homology to B7. B70 is expressed on resting monocytes and dendritic cells and on activated, but not resting, T, NK and B lymphocytes. IT2 substantially inhibited the binding of a CTLA4-immunoglobulin fusion protein to human B-lymphoblastoid cell lines and, together with anti-B7 antibody, completely blocked CTLA-4 binding. Further IT2 efficiently inhibited primary allogeneic mixed lymphocyte responses. These findings indicate that B70 is a second ligand for CD28 and CTLA-4 and may play an important role for costimulation of T cells in a primary immune response.

No abstract

Cervical cancer is a leading cause of death due to cancer among women worldwide. Using transgenic mice to dissect the contributions of the human papillomavirus (HPV) 16 E6 and E7 oncogenes in cervical cancer, E7 was identified previously to be the dominant oncogene. Specifically, when treated with exogenous estrogen for 6 months, E7 transgenic mice developed cancer throughout the reproductive tract, but E6 transgenic mice did not. E6 contributed to carcinogenesis of the reproductive tract, as E6/E7 double transgenic mice treated for 6 months with estrogen developed larger cancers than E7 transgenic mice. In the current study, we investigated whether the E6 oncogene alone could cooperate with estrogen to induce cervical cancer after an extended estrogen treatment period of 9 months. We found that the E6 oncogene synergizes with estrogen to induce cervical cancer after 9 months, indicating that E6 has a weaker but detectable oncogenic potential in the reproductive tract compared with the E7 oncogene. Using transgenic mice that express mutant forms of HPV16 E6, we determined that the interactions of E6 with cellular A-helix and PDZ partners correlate with its ability to induce cervical carcinogenesis. In analyzing the tumors arising in E6 transgenic mice, we learned that E6 induces expression of the E2F-responsive genes, Mcm7 and cyclin E, in the absence of the E7 oncogene. E6 also prevented the expression of p16 in tumors of the reproductive tract through a mechanism mediated by the interaction of E6 with

The lymphocyte surface CD8 antigen is a heterodimer with each chain containing a single Ig-related domain, a hinge-like sequence, a transmembrane segment, and a short cytoplasmic sequence. A soluble form of the rat CD8 alpha chain was produced by introducing a stop codon into the cDNA at the end of the region encoding the extracellular sequence and expressed in Chinese hamster ovary cells. sCD8 alpha was produced at 20 mg/l, and consisted of monomers, dimers, and higher aggregates. The latter could be minimized, but not eliminated, by removal of one of the two cysteine residues in the hinge region by mutation and by growth in serum-free medium. The positions of the N- and O-linked glycosylation sites and the disulphide bond in the Ig-like domain were determined. The MRC OX-8 antibody was shown to react with a region from the CD8 alpha hinge containing 24 amino acids and the antigenic determinant was sensitive to neuraminidase digestion. A construct encoding the Ig-like domain of rat CD8 alpha without the hinge was not expressed in CHO cells, indicating the importance of the hinge region for expression. It seemed possible that the CD8 alpha hinge might facilitate expression of other Ig-related domains and such expression could be detected using the MRC OX-8 antibody. To test the system cDNA constructs were made with the rat CD8 alpha hinge spliced to the V-like domain of mouse CD8 alpha, to the V alpha and V beta domains of a T lymphocyte antigen receptor, and to one or both of the Ig-like domains of the MRC OX-47 membrane antigen. All these forms were expressed as soluble proteins that were detected with the MRC OX-8 antibody. This method may prove useful for the expression of Ig superfamily domains for raising antibodies and other studies.

Sttmmal'yThe adhesion interaction between the immunoglobulin superfamily molecules CD2 and CD58 (lymphocyte function-associated antigen 3) plays an important role in T cell and natural killer cell interaction with various antigen-presenting and target cells. Determination of the solution structure of rat CD2 domain 1 has allowed a model of human CD2 domain 1 to be generated, and a series of mutants based on this model have been made. Residues of domain 1 of human CD2 predicted to be solvent exposed were substituted with the equivalent residues present in the rat CD2 molecule. The ability of these mutants to mediate rosetting with human and sheep erythrocytes was studied. Results show that the binding site of CD2 for both human and sheep CD58 maps to the tS sheet containing tS strands CC'C"F and G. Residues K34 and E36 in 13 strand C, R48 and K49 in/3 strand C', and K91 and N92 in the loop connecting 13 strands F and G are shown to be critical in the interaction. The data support the proposition that the interaction between CD2 and CD58 involves the major/3 sheet face of CD2.

It has been proposed that antibodies can mimic the binding of a receptor to its ligand and that anti-idiotype antibodies raised against such antibodies can be used to identify the receptor. A large number of antibodies have been raised against CD4, the receptor on T cells for the envelope glycoprotein gp120 of the human immunodeficiency virus, and the site at which gp120 binds to CD4 has been delineated. It has therefore become possible to contrast the fine specificities of a natural ligand (gp120) and antibodies that interact with the receptor at the same site. Here we report that out of a panel of 225 anti-CD4 antibodies, only one showed fine binding specificity that was broadly like that of gp120, but the evidence was against this being an exact mimic. Thus the data indicate that the production of antibody mimics will occur very rarely or not at all and that the anti-idiotype approach is unlikely to be useful. This contention is supported by a review of the results of attempts to use this approach. Taking strict criteria for success, there is no example for which the anti-idiotype approach has led to the discovery of a previously undescribed receptor or other protein of interest.

Slllnnlal'y CD4 is the primary receptor for the human irnmunodeficiency virus type 1 (HIV-1). Early mutational studies implicated a number of residues of CD4, centered in the region 41-59, in binding to gpl20. However, further mutational analyses, together with studies using inhibitory antibodies or CD4-derived peptides, have suggested that other regions of CD4 are also involved in binding or postbinding events during infection. To resolve these ambiguities, we used rat CD4 mutants in which particular regions were replaced with the corresponding sequence of human CD4. We have previously shown that some of these are able to bind HIV-1 gpl20, and here we test their ability to act as functional receptors. We find that the presence of human CD4 residues 33-62 is enough to confer efficient receptor function to rat CD4, and we conclude that it is unlikely that regions of CD4 outside this sequence are involved in specific interactions with HIV-1 during either infection or syncytium formation.T he CD4 cell surface glycoprotein is the primary receptor for HIV-1, and its high affinity interaction with the viral glycoprotein gp120 is the initial step in HIV-1 infection of most susceptible cells (1, 2). The extracellular region of CD4 contains four Ig superfamily domains, the amino terminal of which closely resembles an Ig light chain variable domain (3, 4). The three-dimensional structure of the amino-terminal two domains has been confirmed by x-ray crystallography (5, 6). In vitro mutational analyses have identified residues 41-59 within the C'-C" ridge of CD4 domain 1 as the probable binding site for gpl20 (5, 7-11).The postbinding events that lead to fusion of the virus and cellular membranes are unclear. CD4 may induce conformational changes in gpl20 that expose the fusogenic region of gp41 (12-15). It has been proposed that regions of CD4 outside the gp120-binding site might be involved in this process (16-18). For example, mAbs that bind the F-G turn (equivalent to the CDR3 loop of Igs) block syncytium formation and viral entry while only poorly inhibiting the binding of gpl20 or HIV-1 (16). Derivatized peptides corresponding to residues 81-92 of human (h) 1 CD4, which encompass this loop, inhibit postbinding events (17). Another study implicated this region by demonstrating that chimpanzee 1 Abbreviations used in this paper: h, human; m, mutant; m.o.i., multiplicity of infection; r, rat.CD4 and hCD4 with a chimpanzee CDR3 loop were unable to mediate HIV-l-dependent syncytium formation, unlike wild-type hCD4 or chimpanzee CD4 with a hCDR3 loop (19).Rat (r)CD4 shares only 50% homology with hCD4 and does not function as a receptor for HIV-1. We have taken advantage of this to determine the regions of CD4 involved in HIV-1 infection by replacing rCD4 sequence with the equivalent hCD4 residues. Using this approach initial studies established that the gp120 binding site of CD4 is encompassed by residues 33-62 of domain 1 (20). In the present study we have taken the analysis a stage further and determined whether regions o...

A partial cDNA clone for mouse leukosialin was isolated by use of a rat leukosialin cDNA probe. The mouse cDNA was then used to isolate genomic clones that corresponded to the two mouse genes detected in Southern blots. One gene encoded an open reading frame for the homologue of rat leukosialin and this gene was notable for the absence of introns within the coding sequence. A lack of introns has previously been observed for the human leukosialin gene (Shelley, C. S., Remold-O'Donnell, E., Rosen, F. S. and Whitehead, A. S., Biochem. J., submitted). The other mouse gene was an intronless pseudogene for a leukosialin-related sequence. The presence of only one functional gene that lacked coding-region introns established that molecular heterogeneity in mouse leukosialin could not arise from multiple genes or alternative splicing of exons. The sequence of mouse leukosialin suggested an extracellular segment with a high content of O-linked carbohydrate, as is the case in the rat and human. In addition the mouse molecule had one possible N-linked glycosylation site. The cytoplasmic domain of 124 amino acids was highly conserved between rodent and human leukosialins for the functional genes but not for the pseudogene. This suggests an important functional role for the cytoplasmic domain.