Association of maternal killer-cell immunoglobulin-like receptors and parental HLA-C genotypes with recurrent miscarriage.

BACKGROUND: The natural killer (NK) cells at the site of placentation express killer-cell immunoglobulin-like receptors (KIR) that can bind to human leukocyte antigen (HLA)-C molecules on trophoblast cells. Both these gene systems are polymorphic and an association of particular maternal KIR/fetal HLA-C genotypes has been shown in pre-eclampsia. Pre-eclampsia and recurrent miscarriage (RM) share the pathogenesis of defective placentation and therefore we have now genotyped couples with RM. METHODS AND RESULTS: DNA was obtained from the male (n = 67) and female (n = 95) partners of couples with three or more spontaneous miscarriages and genotyped for HLA-C groups and 11 KIR genes using the PCR-sequence-specific primer method (SSP). The frequency of the HLA-C2 group was increased in both parents (reaching significance only in the male partners, P = 0.018) compared with a parous control population. The KIR gene frequencies of the male partners were similar to controls, but the women had a high frequency of KIR AA haplotypes that lack activating KIR. In particular, the activating KIR for HLA-C2 groups (KIR2DS1) was significantly lower in these women (P = 0.00035, odds ratio 2.63, confidence interval 1.54-4.49). CONCLUSIONS: This is the first report to identify a genetic male factor that confers risk in RM. These findings support the idea that successful placentation depends on the correct balance of NK cell inhibition and activation in response to trophoblast.

How Does the maternal immune system contribute to the development of pre-eclampsia?

An immunological aura has hovered over the study of pre-eclampsia for many years but there has still been little progress in explaining the various 'immune' phenomena associated with this elusive disease. When considering the primary defect of placentation that leads to pre-eclampsia the focus should be on the intermingling of the invasive placental trophoblast cells with maternal leukocytes in the uterine wall. The MHC status of trophoblast cells is a crucial factor to be considered, as these molecules can act as ligands for uterine immune cells, including T cells, NK cells and myelomonocytic cells. Extravillous trophoblast cells express an unusual combination of HLA-C, HLA-G and HLA-E molecules and only one of these HLA molecules, HLA-C, shows any appreciable polymorphism. In humans, uNK cells express an array of receptors, some of which are known to bind to the HLA class I molecules expressed by extravillous trophoblast cells. HLA-C is the dominant ligand for killer immunoglobulin-like receptors (KIR) expressed by uterine NK cells that may deliver an inhibitory or activating signal. KIR haplotypes comprise two groups, A and B; these differ principally by having additional activating receptors in the B haplotype. In any pregnancy, the maternal KIR genotype could be AA (no activating KIR) or AB/BB (presence of between one and five activating KIRs). The HLA-C ligands for KIR on trophoblast cells may belong to two groups, C1 and C2 that are defined by a dimorphism at position 80 of the alpha1 domain. This maternal-fetal immunological interaction, occurring at the site of placentation, therefore involves two polymorphic gene systems, maternal KIRs and fetal HLA-C molecules. Uterine NK-cell function is thus likely to vary in each pregnancy. In pre-eclamptic pregnancies we have found that some KIR/HLA-C combinations appear unfavourable to trophoblast-cell invasion due to the overall signals that the NK cell receives. The academic excitement of this work is the realisation that this is a novel form of allorecognition based on NK cells that operates entirely differently from self/non-self discrimination used by T cells.

Paternal monoallelic expression of PEG3 in the human placenta.

Genomic imprinting is the phenomenon whereby mono-allelic expression of certain genes occurs depending on their parental origin. The observation that imprinting only occurs in placental mammals has led to the suggestion that it may play a role in this form of reproduction. In the present study we have investigated the pattern of expression of the human PEG3 gene in the early to term placenta, as well as the uterus and ovary, using RT-PCR, northern blot and in situ hybridization. A comparison is made with the expression of Peg3 in the mouse by histochemical staining in betageo knock out mice. We have demonstrated high levels of PEG3 in the human placenta and have localized the signal to the layer of villous cytotrophoblast cells. In contrast, the pattern of expression of Peg3 in the mouse placenta is less restricted, the message being present in all trophoblast populations. Thus, expression of PEG3/Peg3 in the human and mouse placenta is not directly comparable. We have also detected PEG3 message in the ovarian stroma. We have sequenced the human PEG3 gene from exon 3 to exon 9. By utilizing a polymorphism detected in exon 9, we have established that only the paternal allele is expressed in human placenta. Human PEG3 is therefore maternally imprinted as in mouse.

Angiogenic growth factor messenger ribonucleic acids in uterine natural killer cells.

Angiogenesis is essential for endometrial growth and repair, and disruption of this process may lead to common disorders of women, including menorrhagia and endometriosis. In pregnancy, failure of the endometrial spiral arterioles to undergo remodeling leads to preeclampsia. Here we report that in addition to vascular endothelial growth factor A (VEGF-A), human endometrium expresses messenger ribonucleic acids (mRNAs) encoding VEGF-C, placenta growth factor (PlGF), the angiopoietins, angiopoietin 1 (Ang1) and Ang2, and the receptors VEGFR-3 (Flt-4), Tie 1, and Tie 2. Levels of VEGF-C, PlGF, and Tie 2 changed during the menstrual cycle. Intense hybridization for VEGF-C and PlGF mRNAs was found in uterine nature killer cells in secretory phase endometrium and for Ang2 mRNA in the same cells in the late secretory phase. Interleukin-2 (IL-2) and IL-15 up-regulated VEGF-C, but not PlGF or Ang2, mRNA levels in isolated NK cells. Conditioned medium from decidual NK cells did not induce human umbilical vein endothelial cell apoptosis. These results indicate that human endometrium expresses a wide range of angiogenic growth factors and that uterine nature killer cells may play an important role in the abnormal endometrial angiogenesis that underlies a range of disorders affecting women.

HLA-G expression by tumors.

PROBLEM: It has been proposed that the expression of the non-classical MHC class I antigen, HLA-G, by trophoblast is one mechanism by which the placenta evades attack by maternal uterine NK cells. A similar mechanism is thought to be operative in the escape from immunosurveillance by tumor cells. However, data on the expression of HLA-G by tumor cells are highly conflicting. METHOD OF STUDY: In the present study, we have examined tissue sections from a wide variety of tumors by immunohistology and also several cell lines by flow cytometry and RT-PCR. Whilst very faint bands were detected in three cell lines (hepG2, Mead, CaSki) by RT-PCR, no tumors or cell lines were observed to express HLA-G protein. Furthermore, we found that tumor deposits are not usually infiltrated by NK cells. CONCLUSION: Our observations, therefore, do not support the proposal that tumor cells express HLA-G in order to evade host NK cell cytolysis.

HLA-E is expressed on trophoblast and interacts with CD94/NKG2 receptors on decidual NK cells.

Non-classical MHC class I molecule HLA-E is the ligand for CD94/NKG2 NK cell receptors. Surface expression of HLA-E requires binding of specific HLA class I leader sequences. The uterine mucosa in early pregnancy (decidua) is infiltrated by large numbers of NK cells, which are closely associated with placental trophoblast cells. In this study we demonstrate that trophoblast cells express HLA-E on their cell surface in addition to the previously reported expression of HLA-G and HLA-C. Furthermore, we show that the vast majority of decidual NK cells bind to HLA-E tetrameric complexes and this binding is inhibited by mAb to CD94. Thus, recognition of fetal HLA-E by decidual NK cells may play a key role in regulation of placentation. The functional consequences of decidual NK cell interaction were investigated in cytotoxicity assays using polyclonal decidual NK cells. The overall effect of CD94/NKG2 interaction with HLA-E is inhibition of cytotoxicity by decidual NK cells. However, since decidual NK cells are unable to kill trophoblast even in the presence of mAb to MHC class I molecules and NK cell receptors, HLA-E interaction with CD94/NKG2 receptors may regulate other functions besides cytolysis during implantation.

Recognition of trophoblast HLA class I molecules by decidual NK cell receptors--a review.

During placentation the extravillous trophoblast (EVT) cells migrate through the decidua towards the maternal spiral arteries. The walls of the arteries are then destroyed by trophoblast resulting in an increased blood flow to the fetus. These EVT express HLA-G, HLA-E and HLA-C, an unusual combination of two non-classical and one classical MHC class I molecules. The decidua is infiltrated by distinctive uterine natural killer (NK) cells during the time of trophoblast invasion. These cells express a variety of receptors (CD94/NKG2, KIR and ILT) which are known to recognize HLA class I molecules. There is, therefore, a mechanism for molecular recognition of the placental trophoblast cells. The possible functional consequences of this uterine NK cell-trophoblast interactions are uncertain. One possible result is in an altered NK cell cytokine profile which modulates the invasive proclivity of the EVT. In this way placentation could be controlled.

Surface expression of HLA-C antigen by human extravillous trophoblast.

In this paper definitive evidence that the classical class I product, HLA-C, is expressed on the surface of normal trophoblast cells is provided. HLA-C transcripts were sequenced from cDNA isolated from first trimester trophoblast cells obtained by flow cytometric sorting. Both paternal and maternal alleles were transcribed. HLA-C proteins were demonstrated by biochemical analysis and found on the cell surface in association with beta(2)-microglobulin. Upregulation of cell surface HLA-C but not HLA-G expression after interferon (IFN)-gamma treatment was demonstrated by flow cytometric analysis. Immunohistology has confirmed HLA-C is expressed by all extravillous subpopulations in vivo. The question of whether trophoblast HLA-C molecules interact with decidual NK cells expressing killer Ig-like receptors (KIR) has also been addressed. Our results demonstrate that extravillous trophoblast expresses at least two HLA class I molecules, HLA-G and HLA-C on the cell surface.

Human decidual natural killer cells express the receptor for and respond to the cytokine interleukin 15.

The natural killer (NK) cells that are present in the uterine mucosa (decidua) during early pregnancy have a distinctive phenotype, CD56(bright) CD16(-). These cells have previously been shown to proliferate and be activated by interleukin (IL)-2. However, IL-2 is absent from the decidua and placenta, and we have therefore investigated whether IL-15 is present in the uterus and can act on decidual NK cells. Both IL-15 mRNA and protein were found in a variety of cells but particularly in decidual macrophages. IL-15 induced a proliferative response in decidual NK cells that was blocked by anti-IL-15 and was augmented by stem cell factor. The cytolytic activity of decidual NK cells against K562 was augmented. Interestingly, in contrast to IL-2, although activation with IL-15 resulted in some killing of JEG-3 choriocarcinoma cells, normal trophoblast cells remained resistant to lysis. These findings suggest that IL-15 is a candidate cytokine responsible for NK cell proliferation in vivo in the progesterone-dominated secretory endometrium and early decidua.

Molecular studies of trophoblast HLA-G: polymorphism, isoforms, imprinting and expression in preimplantation embryo.

There is considerable interest in human HLA-G arising from the observation that it is expressed selectively on the surface of extravillous trophoblast, the fetal cell population directly in contact with the mother. We investigated several aspects of the molecular biology of this unusual molecule. Limited polymorphism at the nucleotide level, and even more restricted variation at the amino acid level, was found in our Caucasian population. A further unusual aspect of HLA-G is the occurrence of alternatively spliced mRNAs. Spliced messages that could give rise to either membrane-bound or soluble proteins have been reported and six of these alternative forms were detected in all first trimester and term placentae, highly purified villous and extravillous trophoblast and the cell lines, JEG-3 and 221-G. An additional novel splice variant involving loss of part of the 3'-untranslated region was observed with two alleles. Using a sensitive RNase protection assay higher levels of the membrane-bound RNAs as compared to the soluble forms were detected in first trimester and term placentae as well as in JEG-3. Contrary to previous findings our term samples taken from the maternal aspect showed higher levels of both mRNA species when compared to first trimester placenta. The question of imprinting was addressed through the detection of heterozygotes both in placental tissue and, more tellingly, in the purified trophoblast cells. There was no evidence of imprinting. In addition we did not find mRNA for HLA-G in human two to eight-cell embryos or in blastocyst or in sperm samples.

Human uterine lymphocytes.

During the luteal phase and the early months of pregnancy, there is a dense mucosal infiltration of CD56+ natural killer (NK) cells. These uterine NK cells have a phenotype (CD56bright, CD16-, mCD3-) which distinguishes them from peripheral blood NK cells (CD56dim, CD16bright, mCD3-). The uterine NK cells are in close association with extravillous trophoblast (EVT) cells which infiltrate into the decidua and maternal spiral arteries. This subpopulation of trophoblast expresses two human leukocyte antigen (HLA) class I molecules, HLA-G and HLA-C. Circulating NK cells express receptors for HLA class I molecules. We have recently found evidence that similar receptors are present on decidual NK cells belonging to both the Killer Inhibitory Receptor (KIR) and CD94 families. The repertoire of NK receptors expressed varies between different women. The findings indicate that decidual NK cells do have receptors for trophoblast HLA class I molecules. Experiments are underway to determine the effects of this interaction on NK cell function.

Evaluation of trophoblast HLA-G antigen with a specific monoclonal antibody.

A monoclonal antibody to HLA-G has been generated by immunizing HLA-A2.1/human beta 2-microglobulin (beta 2 m) double transgenic mice with murine L cells transfected with both human beta 2 m and HLA-G. This monoclonal antibody, designated as G233, has been found not to cross-react with other HLA class I antigens when tested on numerous cell lines by flow cytometry. With immunohistology, all populations of extravillous trophoblast (cell columns, interstitial trophoblast, endovascular trophoblast, placental bed giant cells) were stained. An extensive range of adult and fetal tissues was also tested but none reacted with monoclonal antibody G233, including those previously reported to express HLA-G mRNA, indicating that the protein has a highly restricted distribution. Failure to detect HLA-G in the fetal thymus raises the question as to how T-cell tolerance to this antigen is induced. Immunoprecipitation of trophoblast surface proteins with monoclonal antibody G233 revealed a heavy chain of 39 kDa and a light chain of 12 kDa, indicating that HLA-G expressed on the surface of trophoblast is complexed with beta 2 m. However, sequential immunoprecipitation with monoclonal antibody W6/32 followed by monoclonal antibody G233 continued to detect a residual band of 39 kDa, suggesting that trophoblast surface HLA-G may also occur as free heavy chains not associated with beta 2 m. Immunoprecipitation followed by two dimensional gel electrophoresis showed that monoclonal antibody G233 recognizes several isoforms of HLA-G from trophoblast similar to the characteristic spot array previously described for HLA-G. This monoclonal antibody G233 will be highly useful in future experiments to elucidate the function of HLA-G.

Uterine NK cells and trophoblast HLA class I molecules.

PROBLEM: To investigate the proposal that NK cells in decidua may control trophoblast migration during implantation of the human placenta. METHOD: Use Mab specific for HLA-G and for HLA-C in association with flow cytometry and immunoprecipitation to determine the expression of these HLA molecules by trophoblast. Expression of Killer inhibitory/activatory receptors (KIR/KAR) and the CD94 receptor by decidual NK cells was also studied. RESULTS: Extravillous trophoblast expressed HLA-G and HLA-C in both beta2m-associated form and as free heavy chains. KIR and KAR are expressed by decidual NK cells. The repertoire of receptors varied between different women and also between blood and decidual NK cells from the same women. The expression of CD94 was also different between blood and decidual NK cells. CONCLUSION: The recognition of HLA-G/HLA-C by KIR/KAR and CD94 could provide a mechanism by which decidual NK cells control trophoblast migration.

Human uterine NK cells have a similar repertoire of killer inhibitory and activatory receptors to those found in blood, as demonstrated by RT-PCR and sequencing.

The expression of natural killer (NK) cell receptors specific for HLA class I molecules has been studied in CD56bright, CD3- NK cells isolated from the pregnant uterine mucosa, the decidua. RT-PCR was performed on cDNA from uterine NK cells with primers designed to amplify members of the killer inhibitory receptor (KIR)/killer activatory receptor (KAR) gene family. Sequencing of the PCR products revealed that uterine NK cells express KIR/KAR which have two or three extracellular immunoglobulin superfamily (Ig-SF) domains. NK receptors for both groups of HLA-C alleles were found. KIR, characterised by a long cytoplasmic tail containing the immune receptor tyrosine-based inhibitory motif (ITIM), and KAR, characterised by a short cytoplasmic domain with a transmembrane region containing a charged lysine, were both identified. Different individuals appear to have a distinct but overlapping repertoire of KIR/KAR. No new members of this NK receptor gene family were identified in the uterine CD56bright NK cells. Similar findings were obtained from non-pregnant endometrial tissues representative of different stages of the menstrual cycle. Immunohistology confirmed that the KIR protein products were expressed by decidual NK cells. These results reveal that NK receptors for trophoblast HLA class I molecules are present in maternal uterine NK cells. Fetal trophoblast cells infiltrating the decidua express HLA-G and HLA-C gene products. This suggests that maternal recognition of the fetus may be mediated by an NK allorecognition system.

Hyperplasia of mouse mammary epithelium induced by expression of the Wnt-1 (int-1) oncogene in reconstituted mammary gland.

We have expressed the Wnt-1 (formerly int-1) oncogene in Balb/c mouse mammary epithelium in vivo, using a tissue reconstitution method in which primary cultures of mammary epithelial cells are infected with a retrovirus vector and then transplanted into mouse mammary fat pads from which the natural epithelium has been removed. Transplants carrying the Wnt-1 gene grew in a hyperplastic pattern, the duct epithelium showing abundant fine side-branches, but without development of clusters of alveoli. The hyperplasias were similar, but not identical, to transplants of normal epithelium in a mid-pregnant host. Transplants of epithelium that expressed Wnt-1 into mammary fat pads of male or ovariectomized females grew to form a similar three-dimensional pattern, but the extent of growth, and so presumably the rate of growth, was slower than in intact females, and there were no terminal end buds at the edges of the outgrowths. Thus, although Wnt-1 may enhance growth of epithelium in the male or ovariectomized-female environment, it does not restore the major mode of growth in the intact female, the extension of major ducts from terminal end buds. Normal epithelium showed no change in morphology when in close proximity to hyperplasia induced by Wnt-1, confirming the limited range of diffusion of Wnt-1 protein in vivo. Our results are consistent with the hypothesis that Wnt-1 acts principally by mimicking the signal that causes ducts to develop side-branches in pregnancy.