IN THIS ISSUE

Engagement of the cellular Toll-like receptor 2 (TLR2) by Candida albicans results in production of proinflammatory cytokines. But the expectation that TLR2^−/− mice would be highly susceptible to infection by the fungus was not borne out in experiments by Netea et al. (p. 3712 ). TLR2^−/− mice survived longer and had a 100-fold lower kidney load of fungi at day 7 following infection with a lethal dose of C. albicans compared with wild-type animals. Intraperitoneal injection of heat-killed C. albicans led to a higher influx of monocytes/macrophages and higher intracellular killing of fungi in TLR2^−/− vs wild-type mice. In vitro stimulation of peritoneal macrophages with heat-killed blastospores resulted in significantly lower levels of IL-10 and increased levels of IFN-γ, but only 20–30% less TNF-α, IL-1β, and IL-6 from TLR2^−/− cells than from control cells. Uninfected TLR2^−/− mice had half the number of CD4⁺CD25⁺ T regulatory (Treg) cells as uninfected wild-type mice. Wild-type Treg cells incubated with a TLR2 ligand had increased survival times. Depletion of Treg cells in TLR2^+/+ mice with an anti-CD25 mAb resulted in a 10-fold decrease of fungal outgrowth in kidneys 7 days postinfection. The data are interpreted to suggest that C. albicans interacts with TLR2 to induce immunosuppression through production of IL-10 by Treg cells.

The TCR complex is down-modulated following ligation with the MHC:peptide complex. Although rapid removal of the ligated TCR is required for continued serial triggering and signal amplification, the molecular mechanisms involved are unknown. La Gruta et al. (p. 3662 ) transfected CD3ζ genes into murine T cell hybridomas lacking endogenous CD3ζ. Expression of the transgene restored TCR expression. A FLAG peptide attached to the NH₂-terminal extracellular portion of one of the transgenes bound more anti-FLAG Ab at 5 h following down-modulation induced by stimulation of T cells with Ag-pulsed B cells than unstimulated transgenic cells. Biotinylated free and TCR-associated CD3ζ were increased in immunoprecipitates from stimulated cells, whereas there was little increase in biotinylation of intact CD3 complex on the cell surface. Mutation of the extracellular domain of CD3ζ at lysine 9, a possible target for biotinylation, did not alter the biotinylation pattern. Pretreatment of transgenic hybridomas with an inhibitor of Src family protein tyrosine kinases decreased TCR down-modulation and abrogated CD3ζ biotinylation. The data indicate that MHC:peptide ligation results in a phosphorylation-dependent architectural change in the TCR:CD3 complex that exposes the CD3ζ NH₂ terminus and leads to dissociation of the CD3ζ dimer. The authors suggest that the change allows T cells to recognize and degrade ligated TCR.

Optimal proliferation and differentiation of B cell receptor (BCR)-stimulated B cells occurs in response to CD40 interacting with its ligand, CD40L, on activated T cells. B cells also can proliferate extensively following CD40-CD40L interactions in the absence of BCR stimulation and minimally following BCR engagement alone. Duddy et al. (p. 3422 ) used an ex vivo system to profile the cytokines released by human B cells under these and other conditions. They found that a 1:15 ratio of CD40L-transfected L cells:B cells was optimal for proliferation. CD40 stimulation alone with the transfected L cells resulted in high levels of IL-10 and IL-6 and minimal amounts of TNF-α and lymphotoxin (LT) compared with unstimulated cells. IL-10 levels were suppressed and TNF-α, LT, and IL-6 levels were increased greatly by BCR and CD40 dual stimulation. Staggered dual stimulation (BCR followed by CD40) eliminated IL-10 and further enhanced production of the other cytokines. Production of TNF-α, LT, and IL-6 was dose-dependent on BCR cross-linking Ab. However, IL-10 secretion was complex. High levels were expressed in response to BCR cross-linking Ab alone; but CD40-stimulated IL-10 production, inhibited by low doses of the Ab, could be amplified by high doses of Ab. The data suggest that B cells play an active role in regulating local immune responses through patterns of B cell cytokines that vary with the type of stimulation.

T cells cross-reactive against ribonucleoprotein and Smith (Sm) Ag components of the spliceosome complex are found in systemic lupus erythematosis patients. However, the molecular basis for this cross-reactivity has not been determined. De Silva-Udawatta et al. (p. 3940 ) selected several human T cell clones reactive with the Sm-B Ag that could be restimulated with the ribonucleoprotein U1–70kD Ag. A high degree of amino acid sequence similarity was found among the CDR3 regions of TCRs cloned from T cells generated against either Ag. Three full-length TCR/A and one TCR/B genes were stably transfected and expressed as A/B pairs in a TCR-deficient Jurkat T cell line. Two transfectants produced IL-2 mRNA when stimulated with either U1–70kD or Sm-B Ag; the third transfectant responded only to the U1–70kD Ag, and a TCR-negative control did not respond to either Ag. The IL-2 response was limited to one Sm-B peptide that spanned the dominant T cell epitope 1 and one nonhomologous U1–70kD peptide that spanned the dominant T cell epitope 3; there was no response to other T cell or control peptides. The experiments demonstrate that the plasticity of the TCR in autoantigen recognition is due to interaction of one TCR with two cross-reactive autoantigenic peptides that lack apparent sequence homology.

Development of irritable bowel disease (IBD)-like colitis in some mice is associated with the interaction of OX40, expressed on CD25⁺CD4⁺ regulatory T (Treg) cells, with its ligand (OX40L); blocking the interaction inhibits IBD. Takeda et al. (p. 3580 ) found that, although OX40 is expressed on Treg cells but not on CD4⁺CD25⁻ T cells, its expression on both types of cells is up-regulated by CD3 stimulation. Thymic Treg cell numbers were higher in mice transgenic for lymphoid expression of OX40L compared with OX40^−/− mice. Treg cells from mice transgenically expressing OX40L suppressed anti-CD3-induced proliferation of wild-type CD4⁺CD25⁻ T cells more effectively than wild-type cells, whereas OX40^−/− Treg cells were less effective. Treg cell anergy to TCR stimulation in vitro was overcome by coincubation with an agonistic anti-OX40 mAb or with APCs from OX40L transgenic mice. Treg cells could not suppress CD4⁺CD25⁻ T cells stimulated in the presence of OX40L APCs or treated with agonistic anti-OX40 mAb if the CD4⁺CD25⁻ T cells expressed CD28. Cotransfer of wild-type Treg and CD4⁺CD25⁻ T cells induced IBD in RAG2 mice transgenically expressing OX40L but not in mice lacking either RAG2 or OX40L and RAG2. OX40^−/−CD4⁺CD25⁻ T cells alone did not induce IBD in either RAG2^−/− mice or in RAG2^−/− mice expressing OX40L. The data indicate that constitutive expression of OX40L potentiates IBD development and that the target of the OX40 signal is the responding T cell population and not the Treg cells.

Although CD93, expressed on myeloid and endothelial cells, platelets, and microglia, is identical with the human receptor for C1 (C1qRp), there is no evidence that C1q and CD93 interact directly. Norsworthy et al. (p. 3406 ) disrupted the mouse CD93 gene by homologous recombination using a targeting vector that generated a deletion within exon 1. CD93^−/− mice were fully viable and fertile, and the maturation and distribution of bone marrow cells was comparable to that of wild-type mice. Thioglycollate-elicited macrophages from wild-type and CD93^−/− mice incubated in vitro with C1q equally phagocytosed RBCs coated with C3 or with IgG2a. RBCs precoated in vitro with IgG1 or IgM before i.p. injection into wild-type and CD93^−/− mice were phagocytosed to the same extent. CD93^−/− mice had a significant decrease in the percentage of macrophages ingesting i.p.-injected apoptotic T cells compared with wild-type controls. Apoptotic cells preopsonized with serum containing C1q were taken up equally by wild-type and CD93^−/− macrophages in vitro. However, both types of macrophages had impaired phagocytic activity for cells preopsonized in serum deficient in C1q. Leukocyte recruitment into the inflammatory peritoneum was comparable in both CD93^−/− and wild-type mice. The authors conclude that CD93 may not be a true C1q receptor but, rather, is involved in phagocytosis of apoptotic cells by peritoneal macrophages in vivo.

Tumor-infiltrating lymphocytes andperipheral blood T cells from patients with renal cell carcinoma (RCC) exhibit defective activation of NF-κB as well as increased susceptibility to apoptosis. Although soluble gangliosides from RCC cells were found to be responsible for these phenomena, the mechanisms involved are unknown. Thornton et al. (p. 3480 ) detected, by supershift assay, two κB binding complexes in Jurkat T cells stimulated in vitro with PMA/ionomycin; the complexes, comprised of RelA and p50 subunits, were absent from activated T cells cocultured with RCC cells but present in activated T cells cocultured with control smooth muscle cells. Overexpression of recombinant RelA reversed both the decrease in intracellular levels of RelA, p50, and an anti-apoptotic protein, Bcl-xl, and the increase in apoptosis following 48 h of coculture with each of three RCC cell lines. Cell lysates of RCC-exposed Jurkat, but not control, cells degraded recombinant RelA protein. T cells cocultured with RCC in the presence of caspase inhibitors were protected from apoptosis and their lysates lost the degradative capability. Pretreatment of RCC with a ganglioside inhibitor before coculture with Jurkat cells resulted in normal levels of the NF-κB protein subunits and one-tenth the amount of apoptosis compared with T cells exposed to untreated RCC. The authors demonstrate that the gangliosides produced by tumor cells induce T cell apoptosis by caspase-mediated degradation of NF-κB subunits.

Summaries written by Dorothy L. Buchhagen, Ph.D.