Young age and a positive family history of colorectal cancer are complementary selection criteria for the identification of Lynch syndrome.

Families at high risk for Lynch syndrome can effectively be recognised by microsatellite instability (MSI) testing. The aim of the present study is to compare the effectiveness of a MSI test for the identification of Lynch syndrome in patients selected by a pathologist mainly based on young age at diagnosis (MSI-testing-indicated-by-a-Pathologist; MIPA), with that of patients selected by a clinical geneticist mainly based on family history (MSI-testing-indicated-by-Family-History; MIFH). Patients with a Lynch syndrome associated tumour were selected using MIPA (n=362) or MIFH (n=887). Germline DNA mutation testing was performed in 171 out of 215 patients (80%) with a MSI positive tumour. MSI was tested positive in 20% of the MIPA-group group compared to 16% in the MIFH-group (P=0.291). In 91 of 171 patients with MSI positive tumours tested for germline mutations were identified as Lynch syndrome patients: 42% in the MIPA-group and 56% in the MIFH-group (P=0.066). Colorectal cancer (CRC) or endometrial cancer (EC) presenting at an age below 50 years would have led to the diagnosis of Lynch syndrome in 89% of these families (CRC below 50 years: 88% and EC below 50 years: 12%). Families detected by MIPA were characterised more often by extracolonic Lynch syndrome associated malignancies, especially EC (P<0.001). Our results indicate that recognition of Lynch syndrome by CRC or EC below 50 years is as effective as a positive family history. Families from patients selected by individual criteria more often harbour extracolonic Lynch syndrome associated malignancies.

Characteristics of small breast and/or ovarian cancer families with germline mutations in BRCA1 and BRCA2.

For families with a small number of cases of breast and/or ovarian cancer, limited data are available to predict the likelihood of genetic predisposition due to mutations in BRCA1 or BRCA2. In 104 families with three or more affected individuals (average 3.8) seeking counselling at family cancer clinics, mutation analysis was performed in the open reading frame of BRCA1 and BRCA2 by the protein truncation test and mutation-specific assays. In 31 of the 104 families tested, mutations were detected (30%). The majority of these mutations (25) occurred in BRCA1. Mutations were detected in 15 out of 25 families (60%) with both breast and ovarian cancer and in 16 out of 79 families (20%) with exclusively cases of breast cancer. Thus, an ovarian cancer case strongly predicted finding a mutation (P < 0.001). Within the group of small breast-cancer-only families, a bilateral breast cancer case or a unilateral breast cancer case diagnosed before age 40 independently predicted finding a BRCA1 or BRCA2 mutation (P = 0.005 and P = 0.02, respectively). Therefore, even small breast/ovarian cancer families with at least one case of ovarian cancer, bilateral breast cancer, or a case of breast cancer diagnosed before age 40, should be referred for mutation screening.

Endothelial cells promote anti-CD3-induced T-cell proliferation via cell-cell contact mediated by LFA-1 and CD2.

T-cell activation requires not only T-cell receptor (TCR) engagement and subsequent TCR/CD3 cross-linking, but also one or more secondary activation signals generated by accessory cells (AC). We investigated the accessory function of endothelial cells (EC) in an in vitro model for T-cell activation where the first cross-linking signal was delivered to peripheral human T lymphocytes by either immobilized anti-CD3 monoclonal antibody (MoAb) or by PHA. In a previous report, we showed that EC provided a potent costimulatory signal in this model system. We have now analysed the nature of the EC-derived costimulatory signal by testing whether EC could be substituted by cytokines, by studying the effect of EC fixation and by testing the involvement of a number of adhesion molecules. Our findings indicate that EC accessory function is mediated mainly by membrane-bound factors. The nature of these membrane-bound factors was analysed by studying the inhibitory properties of a series of MoAbs directed against several adhesion molecules. Antibodies directed against CD44, E-selectin, CD31, CD26, B7/BB1, VLA-4 or VCAM-1 were not inhibitory. However, an inhibition, was clearly observed with antibodies against LFA-1 and CD2. Remarkably, this inhibition was not found with MoAbs to their respective counterstructures ICAM-1 and LFA-3. In summary, we postulate that both LFA-1/ICAM-1, and CD2/LFA-3 interactions are involved in EC accessory function, although the role of the EC-associated adhesion partners is not fully understood.

Characteristics and possible function of endoglin, a TGF-beta binding protein.

Endoglin was recently described as a major TGF-beta binding protein on endothelial cells. In this review we summarize the experimental data on this molecule, discuss some of our own findings, and speculate on its function.

Involvement of LFA-1/ICAM and CD2/LFA-3 in human endothelial cell accessory function.

Interactions of endothelial cells with T cells occur during the process of lymphocyte migration and during leukocyte extravasation in the course of an inflammatory response. Adhesion molecules are crucially involved in these interactions. It has been assumed that, in this way, endothelial cells play an important role in the recruitment of immune cells to sites of inflammation. In vitro experiments indicate that, in addition to this function in leukocyte recirculation, endothelial cells might also be involved as accessory cells in the stimulation of T cells. In this overview, the possible role of adhesion molecules in this process will be discussed.

A new 180-kDa dermal endothelial cell activation antigen: in vitro and in situ characteristics.

PN-E2 is a monoclonal antibody generated against recombinant tumor necrosis factor-alpha (TNF-alpha)-treated human umbilical vein endothelial cells (EC). PN-E2 recognized a molecule with expression levels in vitro that could be downregulated by TNF, and in situ PN-E2 showed only weak reactivity with vascular EC in normal skin, as assessed by immunohistochemical staining. The expression of PN-E2 was considerably increased on EC in various pathologic skin lesions, including psoriasis, granulation tissue, and inflamed skin. PN-E2 antigen expression was analyzed in more detail in vitro on cultured EC and fibroblasts by use of enzyme-linked immunosorbent assay and fluorescence-activated cell sorter techniques. The expression level on human umbilical vein endothelial cells and capillary EC was, in contrast to the in situ immunohistologic findings, invariably high. On fibroblasts, a low expression was found. Incubation of the EC with recombinant TNF-alpha decreased expression by a factor of 2. Incubation of EC with recombinant interferon-gamma resulted in a twofold increase in PN-E2 antigen expression, whereas other cytokines [recombinant interleukin (rIL)-1 alpha, rIL-1 beta, rIL-4, rIL-6], lipopolysaccharide, or recombinant basic fibroblast growth factor had no effect. Immunoelectron microscopy of tissue specimens and EC preparations localized the antigen on the luminal membrane of the endothelium. Immunoprecipitation followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis revealed a major band at 90 kDa and a minor band at 80 kDa under reducing conditions and bands of 180 and 400 kDa under non-reducing conditions. Molecular weight and expression patterns in vitro on EC after incubation with cytokines excluded most of the known endothelium-specific molecules, with the possible exception of endoglin (the 44G4 antigen). We conclude from our findings that this new antigen could be useful as a marker for endothelial activation in skin biopsy material.

The proliferative response of human T cells to allogeneic IFN-gamma-treated endothelial cells is mediated via both CD2/LFA-3 and LFA-1/ICAM-1 and -2 adhesion pathways.

We studied the proliferative response of purified human peripheral blood T lymphocytes (contaminated with less than 0.1% monocytes) to allogeneic MHC class II molecules expressed by endothelial cells (EC) or fibroblasts (FB). In vitro expression of MHC class II molecules was induced by gamma-interferon (IFN-gamma) treatment. The MHC class II expression levels after IFN-gamma treatment on both cell types were comparable. No T cell proliferation was found in the presence of either untreated or IFN-gamma-treated FB, and a marginal proliferation in the presence of untreated EC. IFN-gamma-treated EC, however, were able to induce significant T cell growth. The previously established role of MHC class II molecules in allogeneic T cell proliferation was confirmed in inhibition experiments with monoclonal antibody (mAb) against MHC class II or CD4. In this model, we tested the involvement of a number of adhesion molecules by adding mAbs to cocultures of T cells and IFN-gamma-treated EC. Monoclonal antibodies directed against CD31, CD26, B7/BB1, E-selectin, CD44, VLA-4 alpha-chain and VCAM-1 had no effect, whereas moderate inhibition was observed with anti-VLA-beta-chain and anti-LFA-3. A distinct inhibition of T cell proliferation was observed with mAbs directed against LFA-1, CD2, or a combination of anti-ICAM-1 and -2. Combinations of mAbs directed against T cell adhesion molecules (LFA-1, CD2, VLA-4) or EC adhesion molecules (ICAM-1, and -2, LFA-3, VCAM-1) were able to block T cell proliferation for 100 and 80% respectively. We conclude that CD2/LFA-3 and LFA-1/ICAM interactions are crucially involved in allogeneic T cell/EC interactions.

Accessory function of endothelial cells in anti-CD3-induced T-cell proliferation: synergism with monocytes.

Monoclonal antibodies to CD3 can induce proliferation of resting T cells. In vitro this effect is dependent on the presence of monocytes. They serve as accessory cells providing a co-stimulatory signal after cross-linking of the antibody-coated TcR/CD3 complex by the Fc receptor on the monocytes. We have studied whether endothelial cells can replace monocytes with regard to this function. Highly purified T-cell preparations were cultured in the presence of anti-CD3 antibody, purified monocytes, and human umbilical vein endothelial cells. Anti-CD3 and endothelial cells alone were unable to support T-cell proliferation, due to lack of FcR expression. Addition, however, of as few as 1000 FcR+ monocytes (0.8% of the number of T cells present) to a coculture of T cells and endothelial cells (EC) in the presence of soluble anti-CD3 resulted in a strong proliferation of T cells. When anti-CD3 was presented in an immobilized form (coated to the culture well or to Sepharose beads), or when phytohaemagglutinin was added to the culture as a cross-linking agent, EC could support T-cell proliferation in the absence of any monocytes. We conclude that EC by themselves cannot support the proliferation of pure T cells induced by soluble anti-CD3, but are potent generators of the co-stimulatory signal(s). They provide a suitable starting material to further define this co-stimulatory activity.

Characterization of two Fc receptors for mouse immunoglobulins on human monocytes and cell lines.

We have previously reported a polymorphism in the mitogenic effect of murine (m) IgG1 anti-CD3 monoclonal antibodies. This polymorphism was genetically determined and could be attributed to polymorphism of the Fc receptor (FcR) for mIgG1 present on human monocytes. We have now extended these studies by quantitating FcR expression on monocytes and cell lines by a recently developed EA rosette assay, using the erythrocyte-associated pseudoperoxidase activity. The data show that the polymorphism of the monocyte FcR for mIgG1 is based on a quantitative rather than an absolute difference. Furthermore, this FcR is specific for mIgG1 and does not bind mIgG2a or mIgG2b nor, surprisingly, human IgG. The expression of this FcR on cell lines correlates with their accessory function in IgG1 anti-CD3-induced T cell proliferation. mIgG2a can inhibit the rosetting of monocytes with erythrocytes sensitized with human IgG. The FcR detected by this rosette technique can interact with all four human IgG subclasses but not with mIgG1 or mIgG2b. The expression of this type of FcR on human cell lines correlates well with their ability to support mIgG2a anti-CD3-induced mitogenesis. These direct measurements of FcR expression support the concept that human monocytes have two independent FcR with affinity for mouse IgG: one receptor specific for mIgG2a (which also binds human IgG), and a second specific for mIgG1.

A new photometric method for the quantitation of Fc receptors for murine IgG1 on human monocytes and cell lines.

We have previously shown a polymorphism of human Fc receptors for mouse IgG1 using an EA rosette technique in which human erythrocytes sensitized with a murine IgG1 monoclonal antibody against glycophorin A acted as indicator cells. We now describe a method to quantitate this EA rosetting using the pseudoperoxidase activity present in erythrocytes. This photometric assay allows the sensitive quantitative determination of Fc receptor expression on human monocytes and cell lines. Not only the human Fc receptor for murine IgG1 can be studied in this way, but the method can also be applied to other Fc receptors. An important factor in this type of rosette assay appears to be the amount of negative charge present on the surface of the indicator erythrocytes. Using alcian blue as a probe, we found that this negative charge is higher on human erythrocytes than on sheep erythrocytes, which may contribute to a better signal-to-noise ratio. The method described facilitates the characterization of Fc receptors and permits the rapid screening of monoclonal anti-Fc receptor antibodies.

Polymorphism in mitogenic effect of IgG1 monoclonal antibodies against T3 antigen on human T cells.

It has recently been described that monoclonal antibody OKT 3, that reacts with the T3 determinant on all peripheral T lymphocytes in man, is also mitogenic for T cells. We have produced two monoclonal antibodies (WT 31 and WT 32) which apparently react with the T3 antigen. We have now tested the mitogenic effect of these antibodies and compared it with the mitogenicity of three other anti-T3 monoclonal antibodies, OKT 3, UCHT 1 (ref. 4) and anti-Leu 4 (ref. 5). Two distinct patterns were observed. WT 32 and OKT 3, both of the IgG2a subclass, were mitogenic for human T cells in all cases studied. By contrast, WT 31, UCHT 1 and anti-Leu 4, all of the IgG1 subclass, were devoid of mitogenic effect in 30% of the individuals tested (non-responders). The mitogenicity of all five anti-T3 antibodies was fully dependent on the presence of monocytes. Addition of purified monocytes from a responder to purified lymphocytes from a non-responder induced responsiveness to both IgG1 and IgG2a anti-T3 antibodies. These results suggest that the polymorphism in the mitogenic effect of these IgG1 antibodies is caused by polymorphism in monocyte function, possibly at the level of the Fc receptor that reacts with mouse IgG1.