Heat shock and arsenite induce expression of the nonclassical class I histocompatibility HLA-G gene in tumor cell lines

The nonclassical histocompatibility class I gene HLA-G has a tissue-restricted expression. To explore mechanisms involved in HLA-G transcriptional regulation, we have investigated the effect of stress, including heat shock and arsenite treatment, on HLA-G expression in tumor cell lines. We show that stress induces an increase of the level of the different HLA-G alternative transcripts without affecting other MHC class I HLA-A, -B, -E, and -F transcripts. A heat shock element (HSE) that binds to heat shock factor 1 (HSF1) on stress conditions was further identified within the HLA-G promoter. Considering the ability of HLA-G to modulate the function of immunocompetent cells, we hypothesize a new feature of HLA-G as a signal regulating the immune response to stress.

Use of pattern standard deviation instead of corrected pattern standard deviation in Anderson's criteria

PURPOSE

To determine whether the use of pattern standard deviation instead of corrected pattern standard deviation, as part of Anderson's criteria, makes a difference in categorizing a single field printout using the full threshold strategy on the Humphrey's Field Analyzer.

MATERIALS AND METHODS

Forty-eight patients with glaucomatous field defects of varying severity underwent full threshold perimetry on the Humphrey's Field Analyzer. The third field of one eye was used for the study. The agreement between CPSD and PSD alone was determined. Each field was then categorized using Anderson's criteria first using CPSD and then PSD. The kappa statistic was used to determine agreement in both situations.

RESULTS

The agreement between CPSD and PSD alone, as determined by kappa, was 0.77. The kappa statistic for categorization using Anderson's criteria was 0.82; the weighted kappa was 0.92.

CONCLUSION

There was almost perfect agreement using the PSD instead of CPSD as part of Anderson's criteria. This substitution seems valid for the spectrum of field defects used in this series, at least for the full threshold programs.

The X1 box of HLA-G promoter is a target site for RFX and Sp1 factors

HLA-G gene regulation was investigated with regards to homologies among the pathways regulating both classical MHC class I and MHC class II gene expression. They include four conserved cis-acting regulatory elements located in the proximal promoter region referred to as the W/S/Z box, the X box that is comprised of the X1 and X2 halves, and the Y box with an inverted CCAAT site. The X1 box is the binding site for the ubiquitous RFX complex consisting of three subunits; the X2 box is bound by the X2BP/ATF/CREB family factors. The basic S-X-Y regulatory module interacts with CIITA, which is expressed constitutively in APCs, but may be inducible in others cell types by IFN-gamma. Within HLA-G gene promoter the only conserved motifs are S and X1 boxes. We thus investigated the binding capacity of the HLA-G X box in comparison to that of HLA-DRA and HLA-E. We demonstrate that X2 box mutations in HLA-G promoter affect the binding of ATF/CREB family factors and may privilege the X2 box to access by other shared factors. The X1 box is the target for RFX complex and an additional factor we identified as Sp1. We propose that the X region in the HLA-G gene promoter might participate to the combination of factors which play a role in HLA-G gene activation.

Modulation of HLA-G expression in human thymic and amniotic epithelial cells

Expression of the nonclassical HLA class I antigen, HLA-G, is tightly regulated. HLA-G physiologic expression is mostly restricted to some placental and thymic cell types. Only few established cell lines express HLA-G in vitro. Cytokine-induced expression of HLA-G is hardly observed and also depends on the cell lineage. We assessed expression and cytokine regulation of HLA-G in primary cultures derived from human thymus and amnion epithelial cells, which also express HLA-G in vivo. We show that HLA-G cell surface expression is maintained, but decreases gradually, in primary cultures derived from human thymus and amnion epithelial cells. We also show that IFN-gamma re-induces HLA-G cell surface expression and upregulates classical class I gene expression in both primary cultures and in a thymus derived cell line. We further show that IFN-gamma also upregulates levels of HLA-G transcripts in TEC primary cultures. This study provides evidence that IFN-gamma induction of HLA-G expression occurs in the human amnion and the thymus, and is mediated at the transcriptional level in these tissues. These results also suggest a role for the microenvironment in regulating HLA-G in vivo gene expression in the thymus and amnion membrane.

Identification of HLA-G7 as a new splice variant of the HLA-G mRNA and expression of soluble HLA-G5, -G6, and -G7 transcripts in human transfected cells

The nonclassical HLA-G primary transcript is alternatively spliced to generate several mRNAs that have the capacity to encode four membrane bound isoforms, namely HLA-G1, -G2, -G3, and -G4 and two soluble isoforms HLA-G5 and -G6. We aimed at defining the capacity of full length and truncated soluble HLA-G transcripts to be translated in human cell lines. Our study of HLA-G alternative transcripts in various human tissues led us to identify a new splice variant of the HLA-G mRNA, named G7, in which open reading frame continues in intron 2. Due to the presence of a stop codon within intron 2, HLA-G7 transcripts retain the capacity to be translated as soluble truncated HLA-G proteins bearing the alpha1 domain linked to two specific aminoacids encoded by intron 2. Expression vectors containing cDNAs encoding HLA-G5, -G6, and -G7 isoforms were transfected into human cell lines. The presence of translated HLA-G5, -G6, and -G7 proteins was detected in protein extracts of transfected cells by Western blot and immunoprecipitation, but only the full length HLA-G5 soluble isoform could be clearly detected as a secreted protein in both transfected cells supernatants and body fluids.

HLA-G, -E, -F preworkshop: tools and protocols for analysis of non-classical class I genes transcription and protein expression

Non-classical MHC class I HLA-E, -F, and -G molecules differ from classical class I histocompatibility antigens by specific patterns of transcription, protein expression, and immunological functions. Restriction of the expression pattern of these non-classical antigens may play a key role in modulation of immune responses during pregnancy and diseases but remains to be additionally defined. A specific component of the second International Conference on HLA-G and the 13th HLA-G Histocompatibility Workshop will be dedicated to the analysis of transcription and expression of non-classical class I genes in normal and pathological tissues. In a first step, referred to as the preworkshop, we here report the analysis and conclusions of a working group which was constituted to gather and validate optimal reagents and protocols allowing RT-PCR analysis of HLA-E, -F, -G transcript levels and flow cytometry and immunochemistry analysis of HLA-G expression in cells and tissues. As a result of this work, use of specific primers and probes detecting alternative transcripts of HLA-E, -F, and G have been validated in transfected cells expressing differential pattern of HLA class I antigens. Analysis of the specificity and affinity of collected antibodies has allowed definition of reagents to be proposed for immunochemistry and flow cytometry analysis of HLA-G expression in normal and pathological tissues during the workshop. This work has allowed constitution of an extended workshop group which is now initiating analysis of non-classical class I transcription and expression in various cells and tissues, a collective contribution that will additionally refine our view of the expression of these antigens in normal and pathological situations.

HLA-G promotes immune tolerance

HLA-G is a non-classical major histocompatibility complex class I molecule that differs from the classical HLA-A, -B and -C molecules by (i) alternative splicing of mRNAs encoding for at least four membrane-bound and two soluble HLA-G isoforms, (ii) a limited polymorphism, and (iii) a tissue-restricted distribution. Studies over the past few years have elucidated the function of HLA-G demonstrating inhibition of both NK cell- and T cell-mediated cytolysis. Furthermore, aside from its expression during pregnancy, we have shown that HLA-G is also expressed in solid tumor cells (i.e. human melanoma cell lines and ex vivo melanoma biopsies). Here we present a review of the current state of knowledge of the immunotolerant functions of HLA-G and their implications in materno-fetal tolerance and tumor immunosurveillance.

Metal-assisted unusual hydroxylation at the carbon atom of the triazine ring in dinuclear ruthenium(II) and osmium(II) complexes bridged by 2,4,6-tris(2-pyridyl)-1,3,5-triazine: synthesis, structural characterization, stereochemistry, and electrochemical studies

The reaction of cis-[M(bpy)2Cl2] (M = Ru(II), and Os(II) with 2,4,6-tris(2-pyridyl)-1,3,5-triazine (tptz) in refluxing ethanol-water resulted in the formation of dinuclear complexes of the composition [(M(bpy)2)2(tptz-OH)](PF6)3.nH2O (n = 1 for Ru and n = 0 for Os). In this reaction an unusual metal-induced hydroxylation at the carbon atom of the triazine ring of bridged tptz occurred. However, hydroxylation did not occur in the corresponding mononuclear complexes under similar reaction condition. A comparative study revealed that sufficient electrophilicity on the carbon atom and free movement of the attached pyridyl ring promoted the hydroxylation reaction. The hydroxylated dinuclear complexes exist in two stereoisomeric forms, a rac form (delta delta/lambda lambda) and a meso form (delta lambda/lambda delta). Both diastereoisomers have been isolated in pure form and characterized. The molecular structures of the rac form of Ru(II) complex (3-II) and meso form of the Os(II) complex (4-I) have been established by single-crystal X-ray studies. Crystal data: complex 3-II, monoclinic, C2/c, a = 24.584(7) A, b = 14.309(4) A, c = 41.044(13) A, beta = 92.84(2) degrees, V = 14420.0(7) A3, Z = 8, R = 0.179, wR2 = 0.479; complex 4-I, triclinic, P1, a = 13.444(7) A, b = 14.576(5) A, c = 19.641(7) A, alpha = 98.21(3) degrees, beta = 101.67(4) degrees, gamma = 105.80(4) degrees, V = 3546.0(3) A3, Z = 2, R = 0.093, wR2 = 0.279. The poor data quality of 3-II did not allow anisotropic refinement of non-hydrogen atoms except Ru and P. A PLUTO drawing of this compound is given only to support the molecular structure. 1H NMR data have been used to characterize the diastereoisomers. The dinuclear complexes exhibit unusual electrochemical behavior; cathodic shifts of the metal-centered oxidations and ligand-based first reduction compared to mononuclear complexes have been observed. There is a splitting in the metal-centered oxidation potentials, indicating strong electronic communication between the metal centers. Comproportionation constants (Kcom) of the mixed-valence species have been calculated; the values are in the range 6.03 x 10(4)-4.7 x 10(6). It appears that a metal-metal interaction occurred by an electron-transfer mode across the low-lying pi* orbital of the bridged tptz.

[HLA-G: a tolerance molecule implicated in the escape of tumors from immunosurveillance]

To define rational anti-tumor immunotherapy strategies, the reasons for the frequent lack of efficacy of the immune response to developing tumors must be elucidated. One hypothesis involves expression by tumor cells of molecules with local immuno-suppressive effects. HLA-G is known to be involved in tolerance of the fetus by the maternal immune system. We studied HLA-G expression in primary and metastatic melanomas (ex vivo biopsies and cell lines). We found a high level of HLA-G transcription and expression at the surface of the cells. A variety of patterns of HLA-G isoform transcription and protein expression were seen. The ability of HLA-G to inhibit the cytotoxic effect of two immunocompetent cell types involved in the antitumor response, namely natural killer cells (NK) and T-cells, suggests that HLA-G may help tumors evade the immune system.

Histidine-193 of rat glucosylceramide synthase resides in a UDP-glucose- and inhibitor (D-threo-1-phenyl-2-decanoylamino-3-morpholinopropan-1-ol)-binding region: a biochemical and mutational study

Glucosylceramide synthase (GCS) catalyses the transfer of glucose from UDP-glucose (UDP-Glc) to ceramide to form glucosylceramide, the common precursor of most higher-order glycosphingolipids. Inhibition of GCS activity has been proposed as a possible target of chemotherapeutic agents for a number of diseases, including cancer. Design of new GCS inhibitors with desirable pharmaceutical properties is hampered by lack of knowledge of the secondary structure or catalytic mechanism of the GCS protein. Thus we cloned the rat homologue of GCS to begin studies to identify its catalytic regions. The histidine-modifying agent diethyl pyrocarbonate (DEPC) inhibited recombinant rat GCS expressed in bacteria; this inhibition was rapidly reversible by hydroxylamine and could be diminished by preincubation of GCS with UDP-Glc. These data suggest that DEPC acts on histidine residues within or near the UDP-Glc-binding site of GCS. Mutant proteins were expressed in which the eight histidine residues in GCS were individually replaced by other amino acids. H193A (His193-->Ala) and H193N (His193-->Asn) mutants were unaffected by 0.1 mM DEPC, a concentration that inhibited other histidine mutants and the wild-type enzyme by at least 60%. These results indicate that His193 is the primary target of DEPC and is at, or near, the UDP-Glc-binding site of GCS. His193 mutants were also insensitive to the GCS inhibitor d-threo-1-phenyl-2- decanoylamino-3-morpholinopropan-1-ol, at concentrations which inhibited the wild-type enzyme by >80%. These results have significance for both an understanding of the GCS active site and also for the possible design of new and specific inhibitors of GCS.

HLA-G expression in human melanoma cells: protection from NK cytolysis

Expression of the non-classical HLA-G class I antigen is physiologically restricted to a limited number of tissues including trophoblasts, and is thought to play a role in establishing tolerance of the fetus by the maternal immune system. We investigated whether ectopic expression of HLA-G could also be detected in tumor cells and confer them the ability to escape immune cytotoxic responses. High levels of all alternatively spliced HLA-G transcripts could be detected in melanoma cells by RT-PCR. Analysis of biopsies from a melanoma patient revealed a higher HLA-G transcription level in skin metastasis as compared to healthy skin, while specific amplification of the HLA-G5 transcript was only observable in the tumor. HLA-G protein expression could also be detected in two melanoma cell lines. HLA-G-positive tumors inhibit cytotoxic lysis by the NK cell line YT2C2-PR. This inhibition is not observed with B-EBV cell lines bearing matched class I specificities, and is thought to occur through interaction of HLA-G with inhibitory receptors that are distinct from known KIRs interacting with HLA-E or classical class I molecules. Together, these results confirm that HLA-G expression at the surface of tumor cells can participate in the evasion of antitumoral immune responses and favor tumor progression.

Molecular mechanisms controlling constitutive and IFN-gamma-inducible HLA-G expression in various cell types

HLA-G molecule is thought to play a major role in down-regulating the maternal immune response by inhibiting NK and T cell cytolytic activities. We examined the molecular regulatory mechanisms that may control the restricted expression pattern of the HLA-G gene. We first analyzed protein interactions between nuclear extracts from the HLA-G-positive JEG-3 choriocarcinoma and the HLA-G-negative NK-like YT2C2 cell lines to a 244 bp regulatory element located 1.2 kb from the HLA-G gene, previously shown to direct HLA-G expression in transgenic mouse placenta. This allowed characterization of cell-specific DNA-protein interactions that could account for differential cell-specific expression of the HLA-G gene. In particular two DNA-protein complexes were exclusively observed in YT2C2, suggesting that this HLA-G regulatory element is a target for putative cell-specific repressor factors. We further mapped nuclear factor binding sites to a 70 bp fragment in the upstream region of the regulatory element. We then investigated the effect of IFN-gamma on HLA-G gene expression. HLA-G cell surface expression was enhanced by IFN-gamma treatment in JEG-3 and U937 cell lines and peripheral blood monocytes while no effect was observed in tera-2 teratocarcinoma cell line. HLA-G transcriptional activity was increased only in JEG-3 and U937 cell lines. Activity of the 1.4-kb HLA-G promoter region was unchanged after IFN-gamma treatment in JEG-3 and Tera-2. These results suggest that both post-transcriptional and transcriptional mechanisms implicating IFN-responsive regulatory sequences outside the 1.4 kb-region are involved in IFN-gamma gene activation of the HLA-G gene.

Characterization of 2-[125I]iodomelatonin binding sites in Syrian hamster peripheral organs

The neurohormone melatonin is a key agent in synchronizing the circadian rhythms. At least three types of binding sites have been described for melatonin: the G-coupled, seven-transmembrane domain receptors mt1 and MT2 and a putative binding site called MT3. The latter has been described in hamster brain membranes, and its binding capacity is optimum at 4 degrees C. We further characterized this binding site on other peripheral hamster tissues, including intestine, liver, kidney, lung, muscle, and heart. We found a high level of binding sites (>30 fmol/mg of protein) in intestine and kidney. Furthermore, we completed the existing pharmacological profile of this site, which can now be described as 2-iodomelatonin > 6-chloromelatonin > methy-isobutyl-amiloride > acridine orange > 5-methylcarbonylamino-N-acetyltryptamine > prazosin > N-acetylserotonin > melatonin. This profile was found in all the hamster organs tested that had a large number of binding sites, namely, brain, intestine, kidney and liver. Furthermore, when comparisons were possible, the MT3 pharmacological characteristics were similar to those described in the literature for hamster brain and testis. This profile was compared to the pharmacology obtained on human cloned mt1 and MT2 receptors and proved to be completely different, as expected. We provide new evidence for an alternate melatonin binding site not only in hamster brain but also in some peripheral organs.

Role of HLA-G versus HLA-E on NK function: HLA-G is able to inhibit NK cytolysis by itself

Recent studies have shown that endogenous HLA-E molecules are stabilized on the cell surface upon the expression of HLA-G which contains within its leader sequence, a nonapeptide capable of binding with the HLA-E/beta2m complex. Since HLA-E was found to be the major ligand for the CD94/NKG2A inhibitory receptor, we determined the role of HLA-G versus HLA-E on NK lysis inhibition. We showed that K562 cells transfected with HLA-G1 cDNA are protected from NK lysis by direct interaction between HLA-G1 and killing inhibitory receptor(s). This NK lysis inhibition is not dependent on HLA-E expression, since no HLA-E protein was detected on K562 cells; HLA-G1 is therefore able to inhibit NK lysis by itself.

IL-10 selectively induces HLA-G expression in human trophoblasts and monocytes

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.

Downregulation of HLA class I gene transcription in choriocarcinoma cells is controlled by the proximal promoter element and can be reversed by CIITA

The expression pattern of MHC class I genes in trophoblast cells at the feto-maternal interface is thought to be the basis of the maintenance of pregnancy by protecting the fetus from maternal immune rejection. Transcription of classical HLA class I genes is low or undetectable in most trophoblast cells as well as in JEG-3 and BeWo trophoblast-derived choriocarcinoma cells. The aim of this study was to characterize the regulatory mechanisms that repress HLA-A transcription in these cell types. We show that the -335 to ATG region of the HLA-A11 gene promoter is inactive in JEG-3 and BeWo cells while able to control efficient reporter gene transcription in other cell types, indicating that this region is the target for downregulation of HLA-A expression in choriocarcinoma cell lines. The regulatory sequence involved in HLA-A gene repression was further mapped to a proximal regulatory element within the -107 to +2 ATG region of the promoter. We show that the HLA-A promoter activity cannot be induced by interferon-gamma (IFN-gamma) and that exogenous MHC class II transactivator CIITA is able to induce HLA class I promoter activity in these cells. Stringent transcriptional regulatory mechanisms, implicating the lack of basal and IFN-gamma-inducible class I promoter activity, are thus involved in the downregulation of HLA-A expression in JEG-3 and BeWo trophoblast-derived choriocarcinoma cells.

Heterogeneity of HLA-G gene transcription and protein expression in malignant melanoma biopsies

Nonclassical MHC class I HLA-G antigen expression is tissue specific and is thought to play a role in tolerance of the semiallogeneic fetus by the maternal immune system. Ectopic expression of HLA-G by tumor cells provides them with an additional mechanism of escape from immunosurveillance by host cytotoxic effector mechanisms. The aim of this study was to assess the expression of nonclassical HLA-G antigens in ex vivo human melanoma biopsies. HLA-G mRNA levels corresponding to both membrane-bound and soluble protein isoforms were analyzed in tumor specimens obtained from primary or metastatic melanomas of 23 patients. High levels of HLA-G transcription were detected in tumor specimens in 5 of 23 patients and found to be comparable in both lymph node and skin metastases. HLA-G mRNA transcript levels at tumor sites in 18 of these patients were compared with those in samples of their own healthy skin and were higher in the tumor tissue in 12 patients. Differential expression of mRNA transcripts corresponding to soluble and membrane-bound HLA-G was also observed in some tumor biopsies. HLA-G protein expression was detected in tumors that exhibited high levels of HLA-G transcription by immunofluorescence of frozen sections and Western blot analysis of both tumor and healthy skin biopsies, using anti-HLA-G-specific monoclonal antibodies. This work provides evidence that HLA-G gene transcription and protein expression can be up-regulated ex vivo in melanoma. Our finding that several of the tumors studied expressed high levels of HLA-G provides additional clues as to how a tumor can be selected in vivo to escape from cytotoxic antitumor responses, constituting a new parameter to be considered in the design of therapeutic approaches aimed at enhancing antitumor immune responses.

HLA-G expression: immune privilege for tumour cells?

The non-classical class I antigen HLA-G has been shown to play a role in foeto-maternal tolerance. It interacts with inhibitory receptors to down-regulate natural killer and T cell cytotoxic functions. We here report our investigations on HLA-G expression in melanoma cells and its implication in the induction of immune tolerance to tumours. We provide the first evidence that both malignant human melanoma cell lines and ex vivo biopsies can exhibit high levels of HLA-G transcription with differential HLA-G isoform transcription and protein expression patterns. We further demonstrated that melanoma cells that express HLA-G inhibit NK cytolysis. We thus propose that HLA-G expression may impede the elimination of malignant cells by anti-tumour immune effector cells, constituting a newly described mechanism by which tumour cells may evade immunosurveillance.