I. Ishida

Ligand specificity of a murine gammadelta T-cell receptor-expressing hybridoma (KN6) derived from adult thymocytes has been analyzed in detail. The molecule recognized by the KN6 gammadelta T-cell receptor is expressed on syngeneic cells of various sources (peritoneal macrophages, thymocytes, spleen cells, and Abelson murine leukemia virus-transformed cell lines) and on transformed cells arrested at an early stage of development (e.g., PCC3 embryonal carcinoma cells). Linkage of the gene coding for the KN6 ligand to the major histocompatibility complex genes could be demonstrated by testing KN6 hybridoma reactivity to cells from congenic strains that differ only at H-2. In addition, analysis of recombinant strains indicates that the gene controlling the KN6 ligand is located in or distal to the TL region. Involvement of the KN6 gammadelta T-cell receptor in this recognition process could be directly demonstrated by transferring the KN6 TL specificity after introduction of the productively rearranged KN6 gamma and delta genes into an alphabeta T-cell clone or into the germ line in transgenic mice. These observations raise the possibility that at least some gammadelta cells regulate hemopoietic cell maturation and activation.

A T lymphocyte expresses on its surface one of two types of antigen receptor, T-cell receptor alpha beta or T-cell receptor gamma delta, encoded by a pair of somatically rearranged alpha and beta or gamma and delta genes. It has been suggested that alpha beta T cells are generated only from precursor T cells that failed to rearrange gamma and delta genes in a functional form. However, we found that transgenic mice constructed with functionally rearranged gamma and delta genes produce a normal number of alpha beta T cells. The transgene gamma present in these alpha beta T cells is repressed apparently through an associated cis DNA element (silencer). We propose that some T-cell precursors are committed to generate alpha beta T cells independent of the rearrangement status of their gamma gene and that this commitment involves activation of a factor(s) that interacts with the gamma gene-associated silencer.

To assess the diversity of gamma delta receptors expressed on adult thymocytes we characterized the gamma- and delta-genes used in a panel of T cell hybridomas expressing a TCR-gamma delta-CD3 complex on the cell surface. DNA cloning and sequencing analysis were necessary to determine the used genes because of the presence of multiple rearrangements in these hybridomas. Among eight hybridomas analyzed, five used V gamma 4, two used V gamma 7, and one used V gamma 6 for gamma-genes, and four used V delta 5, two used V delta 4, and two used V delta 7 for delta-genes. The last delta-gene is documented for the first time. V gamma and V delta appear to pair randomly only restricted by the frequency of the utilization of individual V gamma and V delta segments. These gamma- and delta-genes contained N region addition between the V and J region in most cases. Both D delta 1 and D delta 2 regions were used in the most delta-genes as was seen previously. These results show that certain gamma- and delta-gene segments that are rarely used for the expression on fetal thymocytes are preferentially used by adult thymocytes. Also the repertoire of the gamma delta receptors on adult thymocytes is much greater than that on fetal thymocytes. BW5147 specific V4-C1 gamma-gene was expressed in all of the adult hybridomas. Inasmuch as this gene was not expressed in BW5147 cells or in fetal thymocyte hybridomas expressing V5 or V6 gamma genes, there exists a novel V gene segment-dependent control of transcription in the gamma-gene system.

gamma/delta T cells with different TCR repertoires are compartmentalized in different epithelia. This raises the possibility that the TCR-gamma/delta directs homing of T cells to these epithelia. Alternatively, the signals that induce TCR-gamma/delta expression in developing T cells may also induce homing properties in such cells, presumably in the form of cell surface receptors. We have examined this issue by studying the homing of gamma/delta T cells in transgenic mice constructed with specific pairs of rearranged gamma and delta genes. In such mice, most gamma/delta T cells express the transgene-encoded TCR. We find that homing to both skin and gut epithelia is a property of T cells and is not determined by the type of gamma and delta genes used to encode their TCR. We also studied the effect of TCR replacement on the expression of Thy-1 and CD8 proteins on the gamma/delta T cells associated with gut epithelia. Our results show that the expression of the appropriate type of TCR-gamma/delta is not required for the Thy-1 expression by these T cells, suggesting that Thy-1 is not an activation marker. In contrast, CD8 expression by gut gamma/delta T cells seems to depend on the expression of the appropriate type of TCR.

CD100/Sema4D belongs to the semaphorin family, factors known to act as repulsive cues for axons during neuronal development. Mouse CD100 plays a crucial role in both humoral and cellular immunity through ligation of the lymphocyte receptor, CD72. It remains controversial, however, whether human CD100 can function through human CD72 in a manner similar to mouse CD100. To determine the function of human CD100, we generated a recombinant soluble human CD100 protein comprised of the extracellular region of human CD100 fused to the human IgG1 Fc region (hCD100-Fc). hCD100-Fc specifically binds to cells expressing human CD72. As observed previously in the mouse, hCD100-Fc induces the tyrosine dephosphorylation of human CD72, leading to the dissociation of SHP-1 from the CD72 cytoplasmic tail. Consistent with findings for mouse CD100, hCD100-Fc exerts a co-stimulatory effect on B cells and dendritic cells that are stimulated with anti-CD40 mAb. Furthermore, both hCD100-Fc and anti-human CD72 agonistic mAb induce the production of the pro-inflammatory cytokines tumor necrosis factor-alpha, IL-6 and IL-8, even in the absence of anti-CD40 mAb. Collectively, our findings not only demonstrate that human CD100, interacting with human CD72, can function as a ligand in a manner similar to mouse CD100, but also suggest the involvement of human CD100 in inflammatory responses.