To identify the ligand for the B cell-associated antigen CD40, we constructed a chimeric immunoglobulin molecule where the extracellular portion of the CD40 protein replaced the normal immunoglobulin variable region. No binding was detected on resting peripheral blood T cells. However, following T cell activation with phorbol esters and ionomycin, the chimeric protein bound specifically to activated human T cells and precipitated a 35-kDa protein from such cells. The induction of the CD40 ligand was detectable on the cell surface after 1 h, with maximal expression after 8 h of stimulation. The T cells expressing CD40 ligand were predominantly CD4 positive, although a proportion of CD8-positive cells also expressed the protein. There was no particular correlation with CD45 phenotype. Finally, we found that soluble CD40 inhibited T-dependent B cell proliferation. The results are discussed in the context of cognate interactions between B and T cells.

SummaryWe have constructed a soluble chimeric fusion protein between the mouse CD8ot chain and the mouse CD40 T cell ligand. This protein binds to both human and mouse B cells. By itself it induced a modest degree of B cell proliferation, but together with antiimmunoglobulin (antiIg) antibody it greatly stimulated B cell proliferation, as determined by both [3H]thymidine uptake and increase in cell numbers. These data are evidence that the CD40 ligand on T cells provides a signal that drives B cell proliferation. This signal is synergistic with that delivered by anti-Ig antibody.

SumlnaryThis report outlines the B cell phenotype of transgenic mice that overexpresses the mouse CTLA-4-humanyl (mCTLA4-H'y1) protein. Despite the fact that these mice prime CD4 + T cells (Ronchese, F., B. Housemann, S. Hubele, and P. Lane. 1994. J. Exp. Med. 179:809), antibody responses to T-dependent antigens are severely impaired. In contrast, T-independent responses are normal which suggests mCTLA4-H'y1 does not act directly on B cells, but acts indirectly by impairing T cell help. The impaired antibody defect is associated with impaired class switching, with low total immunoglobulin (Ig)G and antigen-specific IgG responses, and an absence of germinal center formation in spleen and lymph nodes but not gut-associated tissues. The defective germinal center formation is associated with a reduction in the degree of somatic mutation in hybridomas made from transgenic mice in comparison with those made from normal mice. It seems likely that mCTLA4-H'y1 exerts its effect by blocking an interaction between T and B cells that induce T cell help for B cells.

SunllTl~lry CD4 + T cell responses were analyzed in transgenic mice expressing a soluble form of murine CTLA-4, mCTLA4-H3,1, which blocks the interaction of the T cell activation molecules CD28 and CTLA-4 with their costimulatory ligands. Consistent with previous reports (Linsley, P. S., P. M. Wallace, J. Johnson, M. G. Gibson, J. L. Greene, J. A. Ledbetter, C. Singh, and M. A. Tepper. 1992. Science (Wash. DC). 257:792), T cell-dependent antibody production was profoundly inhibited in mCTLA4-H'y1 transgenic mice immunized with a protein antigen. Surprisingly, however, transgenic mice could generate quantitatively and qualitatively normal primary T cell responses, as measured by limiting dilution assays and lymphokine production. In addition, in vi r expansion of antigen-specific T cells after secondary or tertiary immunization was enhanced in mCTLA4-H3'I transgenics as compared with normal mice. Although unable to deliver cognate help to B cells in vivo, T cells from mCTLA4-H'y1 transgenic mice were not anergic as they could help B cells to produce specific antibodies when adoptively transferred into nude hosts. Taken together, these data suggest that the engagement of CD28 and/or CTLA-4 may not be required for the induction of T cell responses, as is currently understood, but rather for the expression of T cell effector function such as the delivery of T cell help to B cells.

In mouse Peyer's patches (PP), dendritic cells (DC) are localized in T cell areas as NLDC145+ CD11c+ cells, and in the dome and corona region of the follicle as NLDC145- CD11c+ cells, respectively, suggesting the presence of two different DC populations with distinct roles in antigen uptake, processing, and presentation. However, it is not clear how this relates to DC maturation. In this report, we demonstrate that freshly-isolated CD11c+ DC have the properties of immature DC since they endocytose soluble antigens, phagocytose particulate material such as latex beads, synthetize major histocompatibility complex (MHC) class II and invariant chain, but, at the same time, display low stimulatory activity for resting T cells, as shown in mixed-lymphocyte reaction and oxidative mitogenesis assays. When cultured for 24 h in the presence of the cytokines granulocyte-macrophage colony-stimulating factor and tumor necrosis factor or anti-CD40, the cells undergo dramatic phenotypic and functional changes characteristic of DC maturation. After 24 h stimulation in vitro, CD11c+ cells lose the ability to take up proteins such as ovalbumin, and in parallel with this decline, the biosynthesis of MHC class II and invariant chain is dramatically down-regulated or eliminated. On the other hand cells treated in vitro exhibit on the cell surface higher levels of MHC class II, of co-stimulatory molecules (CD80, CD86), of adhesion molecules (CD44, intercellular adhesion molecule-1), and acquire expression of the interdigitating DC surface marker NLDC145. Concomitantly, the ability to stimulate naive T cells drastically increased after in vitro treatment with both stimuli. Taken together, our results indicate that the majority of DC in the PP are immature in terms of their antigen-uptake capacity. These sentinel antigen presenting cells are strategically positioned at the dome region of PP, where antigens are transcytosed via the M cells from the gut lumen. A second population of mature interdigitating NLDC145+ CD11c+ DC stimulates naive unprimed T cells in interfollicular areas by up-regulation of surface ligands and accessory signals.