Search

Co-stimulatory signals are absolutely required for T cell activation after TCR-MHC-peptide interaction. The most important co-stimulatory signal known so far is mediated by the interaction of CD28 on T cells with B7 on APC. Here we demonstrate that the co-stimulatory signal from the B7 molecule does not necessarily have to come from the same cell which presents antigen. Titration curves obtained by limiting the amount of anti-CD3 mAb suggests that the same amount of TCR-CD3 cross-linking is required for full T cell activation whether B7 is present on the same or on another cell, but that the kinetics of T cell activation is slower when B7 is present on a separate cell from the primary signal. Finally and most importantly we also show that CD45RO+ memory T cells, but not CD45RA+ naive T cells, can be efficiently activated when B7 is expressed on bystander cells. These findings imply that co-stimulatory activation requirements of B7 are more stringent for naive than for memory T cells, which could be an important mechanism involved in the maintenance of self-tolerance.

Monoclonal antibodies RNL-2 and RNL-3 were previously shown to react with four 35-45-kDa proteins, expressed only in small cell lung carcinoma NCI-H82 cells, but to stain a subset of neuroendocrine tissues and neoplasms (Broers, J. L. V., Mijnheere, E. P., Klein Rot, M., Schaart, G., Sijlmans, A., Boerman, O. C., and Ramaekers, F. C. S. (1991) Cancer 67, 619-633). We used RNL-2 and RNL-3 to isolate cDNA sequences that code for proteins containing the two corresponding epitopes and utilized such cDNAs to develop second generation antibodies. Using these antibodies, we identified a novel 135-kDa protein. The corresponding cDNAs were found to belong to a previously unknown gene with a neuroendocrine-specific expression pattern, tentatively designated NSP gene. NSP transcription appeared to result in mRNAs of 3.4 and 1.8 kilobases (kb). In the NCI-H82 cells only, an apparently aberrant transcript of 2.3 kb was found. cDNAs containing coding sequences of the 3.4-, 2.3-, and 1.8-kb transcripts were isolated, and nucleotide sequence analysis revealed extensive sequence overlap and open reading frames for proteins of 776, 356, and 208 amino acids, respectively. The three deduced proteins all have a common carboxyl-terminal part with two large hydrophobic regions. Transfection of the complete coding sequences of the 3.4-kb transcript resulted in the production of a protein with an apparent molecular mass of 135 kDa. This protein is predicted to be highly negatively charged (calculated pI of 4.35), to be rich in proline and serine, and to contain multiple potential phosphorylation sites.

CD5 is a 67-kDa glycoprotein expressed on the cell surface membrane of all T lymphocytes and on a small proportion of B lymphocytes. The physiologic role of this Ag is still unknown. Structural and functional studies of CD5 suggest that it might act as a receptor for a positive signal. CD5-specific mAb augment CD3- or mitogen-induced T cell proliferation, IL-2 secretion, and IL-2R expression and induce a rise in intracellular [Ca2+]. In this report, we describe the purification of mouse CD5 protein (mCD5) and its use as a probe to search for the ligand of CD5. We demonstrate that mCD5 specifically interacts with the mouse B cell differentiation Ag CD72/Lyb-2. Three serologically defined allelic forms of mouse CD72/Lyb-2 can all interact with mCD5. We further show that mCD5 can interact with human CD72/Lyb-2, and similarly, that human CD5 can interact with mouse CD72/Lyb-2. These studies may have major implications for the mechanisms of T-B cell communication.

The glycoprotein CD5 is expressed on the surface membrane of all mature T cells and a small proportion of B lymphocytes. Its exact role in immune interactions is still unknown. Studies indicate that CD5 functions both in mice and humans as a receptor, delivering co-stimulatory signals to T cells in a manner similar to CD2 (ref. 11) and CD28 (ref. 12). Anti-CD5 antibodies stimulate both T-cell proliferation mediated by CD3 in association with the T-cell receptor and secretion of interleukin-2 and expression of its receptor, as well as inducing an increase in intracellular Ca2+ concentration (refs 5-10). To identify the ligand for CD5 we purified the human CD5 protein, labelled it with biotin and used it as a probe. Here we report that CD5 specifically interacts with the cell-surface protein CD72 exclusive to B cells. This interaction is blocked by anti-CD72 antibodies, but not by any other anti-B-cell antibodies. Moreover, non-B cells (mouse L-cell fibroblasts and human Jurkat T cells) expressing a transfected human CD72 complementary DNA could bind to the CD5-biotin conjugate. The results demonstrate that the B-cell surface protein CD72 (Lyb-2 in mice) is the ligand for CD5.