Search

It is now acknowledged that the pattern of HLA-G expression is not restricted to extravillous cytotrophoblast cells, as several studies described HLA-G in HLA class I⁺ cells, such as thymic epithelial cells, cytokine-activated monocytes and some tumors. In these situations, HLA-G may provide an additional inhibitory signal to escape from NK cell-mediated cytotoxicity. Accordingly, the aim of this study was to define the behavior of HLA-G once it is co-expressed into an HLA-A, -B, -C and -E⁺ cell line. For this purpose, HLA-G1 cDNA was transfected into an HLA class I⁺ melanoma cell line which was used as a target towards freshly isolated peripheral blood NK cells. Cytotoxic experiments using either anti-HLA-G1 or anti-HLA-G1 inhibitory receptor mAb show that HLA-G1 boosts the HLA class I-mediated inhibition of polyclonal NK cells through interaction with ILT-2, which appears as the major HLA-G1 inhibitory receptor involved. Nevertheless, HLA-G1 is also able to inhibit the cytolytic activity of an ILT-2^- NK clone which otherwise expresses another HLA-G1 inhibitory receptor belonging to the KIR103 gene family. In order to more precisely define the relative role exerted by HLA-G1 versus -E on polyclonal NK cells, antibody-blocking assays were carried out using either anti-HLA class I or anti-CD94/NKG2A. Results demonstrate that in the absence of HLA-G1, the naturally expressed HLA class I-mediated NK inhibition is predominantly exerted by HLA-E through binding with CD94/NKG2A. In contrast, once HLA-G1 is expressed, it becomes the major NK inhibitory ligand.

In the present study, we demonstrate that the non-classical MHC class I molecule HLA-G impairs specific cytolytic T cell functions in addition to its well-established inhibition of NK lysis. The antigen-specific cytotoxic T lymphocyte (CTL) response analyzed was mediated by CD8⁺ T cells specific for the influenza virus matrix epitope, M58-66, presented by HLA-A2. The transfection of HLA-G1 cDNA in target cells carrying the M58-66 epitope reduced their lysis by these virus-specific CTL. This HLA-G-mediated inhibition of antigen-specific CTL lysis was (i) peptide dose dependent, (ii) reversed by blocking HLA-G with a specific mAb and (iii) still observed despite the blockade of HLA-E/CD94/NKG2A interaction. By inhibiting both CTL and NK functions, HLA-G appears to have an extensive role in immune tolerance.

HLA-G plays an essential role in feto-maternal tolerance by inhibiting lysis by maternal NK cells. The factors that allow tissue-specific activation of HLA-G gene expression in trophoblasts remain to be characterized. We investigated the potential effect of IL-10, a cytokine which is secreted in placenta, on HLA-G gene transcription in trophoblasts. Using Northern blot, RNase protection assay and RT-PCR analysis, we demonstrated that IL-10 enhances steady-state levels of HLA-G transcription in cultured trophoblast cells. We further tested the effect of IL-10 on HLA-G gene transcription and protein expression in peripheral blood monocytes, showing that IL-10 can up-regulate HLA-G cell surface expression in this cell type. This effect of IL-10 is selective, since classical MHC class I products and MHC class II are down-regulated in monocytes following IL-10 treatment. Induction of HLA-G expression by IL-10 on monocytes may thus play a role in down-regulation of the immune response. We propose that IL-10 secretion by trophoblasts during pregnancy may also influence the HLA class I expression pattern at the feto-maternal barrier, thus protecting the fetus from rejection. This should be taken into consideration in the design of treatment for pathologies of pregnancy.

The association of paroxysmal nocturnal hemoglobinuria (PNH) and aplastic anemia (AA) raises the yet unresolved questions as to whether these two disorders are different forms of the same disease. We compared two groups of patients with respect to cytogenetic features, glycosylphos-phatidylinositol (GPI)-linked protein expression, protein C/ protein S/thrombomodulin/antithrombin III activity, and PIG-Agene expression. The first group consisted of eight patients with PNH (defined as positive Ham and sucrose tests at diagnosis), and the second, 37 patients with AA. Twelve patients with AA later developed a PNH clone. Monoclonal antibodies used to study GPI-linked protein expression (CD14 [on monocytes], CD16 [on neutrophils], CD48 [on lymphocytes and monocytes], CD67 [on neutrophils and eosinophils], and, more recently, CD55, CD58, and CD59 [on erythrocytes]) were also tested on a cohort of 20 normal subjects and five patients with constitutional AA. Ham and sucrose tests were performed on the same day as flow-cytometric analysis. Six of 12 patients with AA, who secondarily developed a PNH clone, had clinical symptoms, while all eight patients with PNH had pancytopenia and/or thrombosis and/or hemolytic anemia. Cytogenetic features were normal in all but two patients. Proteins C and S, thrombomodulin, and antithrombin III levels were within the normal range in patients with PNH and in those with AA (with or without a PNH clone). In patients with PNH, CD16 and CD67 expression were deficient in 78% to 98% of the cells and CD14 in 76% to 100%. By comparison, a GPI-linked defect was detected in 13 patients with AA, affecting a mean of 32% and 33% of CD16/CD67 and CD14 cell populations, respectively. Two of three tested patients with PNH and 1 of 12 patients with AA had a defect in the CD48 lymphocyte population. In a follow-up study of our patient cohort, we used the GPI-linked molecules on granulocytes and monocytes investigated earlier and added the study of CD55, CD58, and CD59 on erythrocytes. Two patients with PNH and 14 with AA were studied for 6 to 13 months after the initial study. Among patients with AA, four in whom no GP1-anchoring defect was detected in the first study had no defect in follow-up studies of all blood-cell subsets (including erythrocytes). Analysis of granulocytes, monocytes, and erythrocytes was performed in 7 of 13 AA patients in whom affected monocytes and granulocytes were previously detected. A GPI-anchoring defect was detected on erythrocytes in five of six. However, in one patient, who had a GPI-anchoring defect on leukocytes, we did not observe a defect on erythrocytes. In three patients without a GPI-anchoring defect after the initial study, a defect was detected on leukocytes only, with persistent negative Ham and sucrose tests in one patient, in all cell subsets in another, and on erythrocytes only in the remaining patient. Finally, PIG-Agene expression, studied by Northern blot analysis, showed that the PIG-Afull-length transcript was barely detectable or absent in four patients with PNH and abnormal in length in one. All nine patients with AA who secondarily developed a PNH clone had a normal-sized PIG-Agene transcript. In conclusion, this study failed to show any significant difference, except for the size of the PNH clone, between PNH arising in patients with previous AA and in patients with de novo PNH.