Cellular co-localization of intron-4 containing mRNA and HLA-G soluble protein in melanoma analyzed by fluorescence in situ hybridization.

HLA-G5, -G6, and -G7 soluble isoforms of the immunosuppressive HLA-G molecule are produced from the splice variants of the primary HLA-G mRNA transcript containing intron-4 that encodes a specific 21 amino acids tail. In particular, HLA-G5 interacts with the inhibitory ILT2/4 and KIR2DL4 receptors that are expressed on immune cells. Acquisition of soluble HLA-G in the microenvironment may turn a HLA-G non-expressing cell into a HLA-G-bearing one. To address the question of how to distinguish cells that express soluble HLA-G generated by alternative splicing from those that have acquired HLA-G, we have developed a method capable of detecting intron-4 containing mRNA and protein in situ simultaneously. M8 melanoma cell line either transfected or not with HLA-G5 cDNA was analyzed by indirect immunofluorescence confocal microscopy using double staining with a HLA-G intron-4 digoxygenin labeled probe along with a monoclonal antibody directed against the 21 amino acid tail. The combined fluorescence in situ hybridization was also used on the HLA-G-positive choricarcinoma cell line JEG-3. This method would be helpful to follow-up bona fide HLA-G expression in a heterogeneous cell population and to elucidate the mechanisms underlying soluble HLA-G mediated immune modulation in physiological conditions such as pregnancy and pathophysiological situations such as cancer.

Hypoxia modulates HLA-G gene expression in tumor cells.

Human leukocyte antigen G (HLA-G) molecules are expressed in cytotrophoblasts and play a key role in maintaining immune tolerance at the maternal-fetal interface. HLA-G expression was also reported in inflammatory diseases, organ transplantation, and malignant tumors. The regulatory mechanisms of HLA-G gene expression differ from those of classical HLA class I genes and are still only partially elucidated. Focusing on tumor cells, we previously demonstrated a tight control of HLA-G gene expression by cis-acting epigenetic mechanisms. In the present study, we hypothesized that these processes are dependent of microenvironment conditions, and more particularly, stress conditions like hypoxia. Cellular response to hypoxia is mainly driven by a key transcription factor, hypoxia-inducible factor 1 (HIF-1), and other factors, such as NF-kappaB, involved in angiogenesis and cell survival. Here we confirmed the influence of hypoxia on HLA-G gene induction in the HLA-G-negative M8 melanoma cell line. Moreover, upon treatment with the hypoxia-mimicking desferrioxamine, we demonstrated a decrease in HLA-G gene expression in melanoma FON and choriocarcinoma JEG-3 cell lines, both expressing constitutively HLA-G. Finally, we demonstrated for the first time that the modulation of HLA-G gene expression is dependent of HIF-1 stabilization and thus might be relevant for the control of HLA-G gene expression in hypoxic tumors.