TY - JOUR
AU - Kikkawa,, Fumitaka
AB - Maternal immune tolerance is required for extravillous trophoblasts (EVTs) to invade the decidua without rejection. Endoplasmic reticulum aminopeptidase-1 (ERAP1) generates human leukocyte antigen (HLA) class I-adapted antigenic peptides, but its function in trophoblasts lacking classical HLA class I molecules remains undetermined. Leukemia inhibitory factor (LIF) is produced from decidua during the implantation period and plays a necessary role in establishing pregnancy. This study is intended to investigate the location and the function of ERAP1 in trophoblastic cells, focusing on LIF. Immunohistochemistry showed strong ERAP1 expression in cultured EVTs. In choriocarcinoma cell lines used as a model for trophoblasts, ERAP1 was expressed more intensively in JEG-3 than BeWo cells. Immunoblot analysis and immunocytochemistry localized ERAP1 to the endoplasmic reticulum (ER) in JEG-3 cells. Flow cytometry with HLA-G antibody to monitor the supply of antigenic peptides presenting to HLA-G in the ER showed that reducing ERAP1 transcripts by RNA interference did not affect cell surface expression of membrane HLA-G1 (mHLA-G1) in JEG-3 cells under basal conditions. In LIF-treated JEG-3 cells, cell surface mHLA-G1 expression was increased along with ERAP1 protein and promoter activities. In contrast to nonstimulated cells, eliminating ERAP1 from LIF-treated JEG-3 cells reduced the cell surface mHLA-G1 expression and soluble HLA-G1 secretion. This study provides the first evidence showing that ERAP1 is localized in the ER of trophoblasts and is involved in regulating cell surface HLA-G expression in the presence of LIF. Consequently, ERAP1 would function to present antigenic peptides to HLA-G in trophoblasts. EXTRAVILLOUS TROPHOBLASTS (EVTs), which are derived from villous cytotrophoblasts, extensively invade the maternal decidua and come into contact with various maternal immune cells, notably natural killer (NK) and cytotoxic T cells. To avoid rejection by the semiallogenic maternal uterine wall and to establish successful embryonic implantation, immunological factors that facilitate immune tolerance are required, including the status of human leukocyte antigen (HLA) in EVT, a principal determinant of allograft rejection (1). EVTs lack classical major histocompatibility complex (MHC) class I molecules such as HLA-A and HLA-B, which serves to prevent recognition by maternal alloreactive T cells, but leaves them susceptible to maternal NK cells (2). On the contrary, they do express a nonclassical MHC class I molecule HLA-G (3). The restricted expression of HLA-G in EVTs and its negligible polymorphism suggest a significant role of HLA-G in protecting EVTs from maternal NK cells (1, 4). The HLA-G family includes four membrane-bound proteins, mHLA-G1 to mHLA-G4, as well as three soluble isoforms, sHLA-G1 to sHLA-G3, which are created through alternative splicing (5). As shown by the evidence that HLA-G without antigen binding fails to exit the endoplasmic reticulum (ER), antigen binding is essential to export fully assembled HLA-G proteins to the cell surface (6). Because mHLA-G2 to mHLA-G4 isoforms lack the domain for peptide binding, mHLA-G1 is the only membrane-bound isoform that can translocate to the cell surface for antigen presentation (7, 8). Similarly, only sHLA-G1 is secreted after antigen binding. Most antigenic peptides presented by MHC class I to leukocytes are generated through protein degradation (9–11). Proteasomes degrade precursor proteins in the cytoplasms into oligopeptides of the correct size or extended on their N terminal as HLA class I-adapted peptides, which are then transported across the ER membranes by the transporter associated with antigen processing (TAP). In the ER, the extended N-terminal residues were removed sequentially to generate final antigenic peptides of the correct size. This step of N-terminal trimming is important and is regarded as rate limiting for efficient class I presentation. The aminopeptidases in charge of this step had been unidentified for a long time. We isolated an aminopeptidase that is analogous to placental leucine aminopeptidase from human adipocyte cDNA library, and called the aminopeptidase adipocyte-derived leucine aminopeptidase (A-LAP) (12). Initially, A-LAP was characterized as a cleaving enzyme for angiotensin II and angiotensin III (13). However, several lines of evidence have indicated that A-LAP is responsible for the N-terminal trimming of antigenic peptides in the ER, which was renamed A-LAP endoplasmic reticulum aminopeptidase-1 (ERAP1) (14–16). Similar to other proteins involved in antigen presentation such as TAP, interferon-γ (IFN-γ), a cytokine known to stimulate antigen presentation, induces ERAP1 expression in HeLa cells (14). Recently, we found that ERAP1 is also expressed immunohistochemically in EVTs (17). It is tempting to speculate that ERAP1 plays an important role in presenting antigenic peptides to HLA-G in trophoblasts, which has never been investigated. Various cytokines are produced from uterine leukocytes at the maternal-fetal interface, which would be closely associated with the immune system (18). Leukemia inhibitory factor (LIF) plays an essential part in embryo implantation (19, 20). LIF is expressed in the endometrium maximally during the implantation period, whereas its receptor is localized in villous and extravillous trophoblasts (21), which suggests the possible involvement of LIF in mediating interactions between maternal decidual leukocytes and trophoblasts. The finding that LIF enhances HLA-G expression in trophoblasts supports this concept (22). Interestingly, ERAP1 promoter region contains several putative signal transducer and activator of transcription-3 (STAT-3)-binding sites (23) through which LIF increases transcriptional activity (24). We therefore hypothesized that LIF might induce ERAP1 expression in trophoblasts, which could generate more HLA-G-binding antigenic peptides. This study was intended to examine whether ERAP1 is involved in antigenic peptide presentation to HLA-G in trophoblasts and whether LIF enhances the immune process. We first examined the expression of ERAP1 in cultured EVTs (normal invasive trophoblasts) and human choriocarcinoma cell lines, JEG-3 and BeWo, which are used widely as a model for trophoblasts (25, 26). Then we investigated changes in the cell surface expression of mHLA-G1 in JEG-3 cells, which reflect the supply of antigenic peptides for assembly with HLA-G in the ER (14, 15), by eliminating ERAP1 using small interfering RNA (siRNA) in the absence or presence of LIF. Materials and Methods Antibodies (Abs) Rabbit anti-ERAP1 Ab was previously established as anti-A-LAP Ab (13). The monoclonal Ab (mAb) 16G1 specific to sHLA-G1 was a gift from Dr. Geraghty (Fred Hutchinson Cancer Research Center, Seattle, WA) (27). This study used mAb MEM-G/9 specific to mHLA-G1 (Abcam, Cambridge, UK), anti-β-actin Ab (Sigma-Aldrich Corp., St. Louis, MO), mAb against the ER retention signal Lys-Asp-Glu-Leu (KDEL) (Stressgen Bioreagents, Victoria, Canada), anti-Grp94 mAb (Stressgen Bioreagents), and anti-TAP1 mAb (Medical Biological Laboratories Co. Ltd., Nagoya, Japan). In addition, we used horseradish peroxidase-conjugated goat antirabbit Ab (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) and phycoerythrin-conjugated goat-antimouse Ab (Immunotech, Luminy, France) as secondary Abs. Human chorionic villous explant culture The ethics committee of Nagoya University Graduate School of Medicine approved this study. Written informed consent was obtained from each woman before clinical sampling. Villous explant culture was established using placental tissues obtained from legal abortions (6–9 wk gestation; n = 5) as reported previously (28). Briefly, placental tissues were washed with DMEM (Invitrogen Life Technologies, Inc., Burlington, Canada) and dissected aseptically to remove decidual tissues. Small fragments of placental villi were teased apart, and villous fragments were placed on a four-well chamber slide (Nalge Nunc International, Naperville, IL) coated with collagen-1 (BD Biosciences, Bedford, MA), followed by incubation in DMEM supplemented with 10% heat-inactivated fetal calf serum, penicillin (100 U/ml), streptomycin (100 μg/ml), and 5% amphotericin B at 37 C in a 5% CO2 atmosphere for 48 h. Cell lines and cultures The human choriocarcinoma cell lines BeWo and JEG-3 were obtained from American Type Culture Collection (Manassas, VA) and grown in RPMI 1640 medium (Sigma-Aldrich Corp.) supplemented with 10% fetal calf serum, penicillin (100 U/ml), and streptomycin (100 μg/ml) at 37 C in a 5% CO2 atmosphere. To examine the effect of IFN-γ on ERAP1 expression, BeWo and JEG-3 cells were treated with 2.5–25 μg/ml human IFN-γ (PeproTech EC Ltd., London, UK) for 48 h. For the experiments to evaluate the effects of LIF, JEG-3 cells were exposed to a given dose of human LIF (Chemicon International, Temecula, CA) for 48 h or as stated otherwise. RT-PCR Total RNA (1 μg) isolated from choriocarcinoma cells using an RNeasy kit (Qiagen, Ontario, Canada) was reverse transcribed with a Ready-to-Go RT-PCR Beads kit (Amersham Biosciences, Piscataway, NJ) according to the manufacturer’s directions in 50-μl reaction volumes. RT-PCR was performed to detect HLA-G as previously described (22), using 1-μl aliquots of RT reaction products and the following primers: forward, 5′-CTGACCCTGACCGAGACCTGG-3′; and reverse, 5′-GTCGCAGCCAATCATCCACTGGAG-3′. The quality and quantity of RNA were evaluated by amplifying human glyceraldehyde-3-phosphate dehydrogenase using primers: forward, 5′-GGGGAGCCAAAAGGGTCATCATCT-3′; and reverse, 5′-GAGGGGCCATCCACAGTCTTCT-3′. Immunocytochemistry JEG-3 cells grown on coverslips were rinsed in PBS and fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. Immunocytochemical staining was performed as described previously (28), using anti-ERAP1 and anti-KDEL Abs. The chamber slides were also immunostained with antihuman CD146 Ab (Chemicon International), anti-cytokeratin-7 Ab, and anti-MEM-G/9 Ab at a 1:400 dilution to examine stromal cell contamination. Flow cytometry For flow cytometry, cells were incubated with the primary mAb MEM-G/9 or control Ab for 30 min on ice, followed by an additional 30-min incubation with the secondary Ab. They were then washed in PBS. Flow cytometric data were acquired using a FACSCalibur (BD Immunocytometry Systems, San Jose, CA) and were analyzed using CellQuest software (BD Biosciences, Mountain View, CA). siRNA siRNA for ERAP1 (AACGUAGUGAUGGGACACCAUdTdT and AUGGUGUCCCAUCACUACGdTdT, 186–208 bp human ERAP1) and mouse topoisomerase I as a control (CCUCAACGAGGACACCACCdTdT and GGUGGUGUCCUCGUUGAGGdTdT, 689–707 bp mouse topoisomerase I) were purchased from Dharmacon (Lafayette, CO) and used as described previously (15). Cells were transfected using Oligofectamine (Invitrogen Life Technologies, Inc., San Diego, CA) according to the manufacturer’s instructions, except that the transfection was repeated after 4 h. Immunoblot analysis Cells were washed with PBS and scraped into a buffer [50 mm HEPES-KOH (pH 7.6), 50 mm potassium acetate, 5 mm magnesium acetate, 1 mm dithiothreitol, 0.5 mm EDTA, and 250 mm sucrose] as described previously (14). After gentle homogenization, the homogenate was centrifuged at 1500 × g for 5 min at 4 C to remove cell debris. The microsomes were separated from the cytosolic fraction by spinning the supernatant at 10,000 × g for 5 min. The cytosolic and microsomal fractions of the total cell lysate were separated by SDS-PAGE. They were subsequently analyzed using each specific Ab. Each blot was reblotted with anti-β-actin Ab or anti-Grp94 Ab to assess the consistency of loading amounts. For sHLA-G1, 40 μl culture supernatant was analyzed using anti-HLA-G 16G1 mAb. The bands were scanned and analyzed using a densitometer. Statistical analyses Data are expressed as the mean ± sd. Because data were not normally distributed, we employed nonparametric statistics. Comparisons between groups were made by Mann-Whitney U test with Bonferroni’s correction. Differences were considered significant at P < 0.05. Results ERAP1 expression in cultured EVTs and choriocarcinoma cells We performed ERAP1 immunocytochemistry using EVTs in explant cultures of human chorionic villi to determine whether EVTs express ERAP1. In addition to well-established EVT markers such as cytokeratin-7 (Fig. 1B) and CD146 (Fig. 1C), cells from the explanted villous tip were positive for HLA-G (Fig. 1D). We detected ERAP1 immunoreactivity exclusively in the cytoplasm of the cells (Fig. 1E). Nonimmune rabbit IgG showed no immunostaining in the cells (Fig. 1A), indicating that HLA-G and ERAP1 were expressed specifically in EVTs. Fig. 1 Open in new tabDownload slide Expression of ERAP1 and HLA-G in cultured EVTs. Human villous tissue was cultured for 5 d, and cells grown from the villous tip were observed. Cultured EVTs showed no staining with normal rabbit IgG (A). Immunocytochemical localization of cytokeratin 7 (B) and CD146 (C), which are markers of human EVTs, is shown. HLA-G (D) and ERAP1 (E) protein were detected in cultured EVTs. The expression of ERAP1 is localized on the cytoplasm. The data shown here are representative of five experiments. Bars, 100 μm. Fig. 1 Open in new tabDownload slide Expression of ERAP1 and HLA-G in cultured EVTs. Human villous tissue was cultured for 5 d, and cells grown from the villous tip were observed. Cultured EVTs showed no staining with normal rabbit IgG (A). Immunocytochemical localization of cytokeratin 7 (B) and CD146 (C), which are markers of human EVTs, is shown. HLA-G (D) and ERAP1 (E) protein were detected in cultured EVTs. The expression of ERAP1 is localized on the cytoplasm. The data shown here are representative of five experiments. Bars, 100 μm. Cells growing out from the explanted villous tip were so fluctuating that we were unable to obtain a sufficient number of cells to perform our study. We therefore used choriocarcinoma cells as a model for normal trophoblasts in additional experiments. First, we examined the expression of ERAP1 in human choriocarcinoma cell lines JEG-3 and BeWo under basal and IFN-γ-stimulated conditions, because one representative characteristic of ERAP1 is IFN-γ dependency. Western blot analyses showed strong ERAP1 signals of the appropriate size in JEG-3 cells, which were induced 2.5-fold by IFN-γ-treatment, whereas BeWo cells had faint signals with little induction by IFN-γ (Fig. 2, A and B). We also assessed HLA-G expression levels in the two cell lines and obtained results similar to those of ERAP1 proteins: more abundant HLA-G in JEG-3 than BeWo cells (Fig. 2C). These results indicated that JEG-3 cells were more suitable to examine the interaction between ERAP1 and HLA-G; for that reason, we used JEG-3 cells in subsequent experiments. Fig. 2 Open in new tabDownload slide Expression of ERAP1 and HLA-G in BeWo and JEG-3 cells. A, Immunoblot analysis of ERAP1 in BeWo and JEG-3 cells with or without IFN-γ-treatment. Total homogenates (80 μg) from BeWo or JEG-3 cells incubated with IFN-γ (0–25 μg/ml) for 48 h were analyzed by immunoblotting with anti-ERAP1 or anti-β-actin Ab. The data shown are representative of three separate experiments with independent triplicate samples. B, Quantitative densitometric analyses of ERAP1 proteins normalized for β-actin (mean ± sd). *, Comparison is significant (P < 0.05). C, RT-PCR of HLA-G using RNA from BeWo and JEG-3 cells. HLA-G mRNA was detected by RT-PCR that specifically amplifies a 338-bp product in both cell lines. The level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression was used as a measure of the relative RNA in each lane. Fig. 2 Open in new tabDownload slide Expression of ERAP1 and HLA-G in BeWo and JEG-3 cells. A, Immunoblot analysis of ERAP1 in BeWo and JEG-3 cells with or without IFN-γ-treatment. Total homogenates (80 μg) from BeWo or JEG-3 cells incubated with IFN-γ (0–25 μg/ml) for 48 h were analyzed by immunoblotting with anti-ERAP1 or anti-β-actin Ab. The data shown are representative of three separate experiments with independent triplicate samples. B, Quantitative densitometric analyses of ERAP1 proteins normalized for β-actin (mean ± sd). *, Comparison is significant (P < 0.05). C, RT-PCR of HLA-G using RNA from BeWo and JEG-3 cells. HLA-G mRNA was detected by RT-PCR that specifically amplifies a 338-bp product in both cell lines. The level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression was used as a measure of the relative RNA in each lane. Subcellular localization of ERAP1 In HeLa cells, ERAP1 was shown to localize in the ER (14). However, whether ERAP1 also exists in the ER of trophoblastic cells, which lack MHC class I molecules, remains unknown. We examined this in two ways. First, we fractionated JEG-3 cell homogenate by differential centrifugation, and performed immunoblot analysis (Fig. 3A). In contrast to the weak signals in the cytosol fraction, the microsomal fraction gave strong signals, the intensity of which was comparable to that of total homogenate, suggesting that ERAP1 was distributed preferentially to the microsomal fraction. Second, we analyzed the ERAP1 location in JEG-3 cells immunocytochemically with anti-ERAP1 Ab (Fig. 3B) and Ab against ER retention signal sequence KDEL (Fig. 3C). With confocal microscopy, merged images showed that ERAP1 immunoreactivity was colocalized with KDEL in vesicular structures in the cytosol of JEG-3 cells (Fig. 3D). These results indicated that ERAP1 was expressed in the ER lumen of JEG-3 cells. Fig. 3 Open in new tabDownload slide Localization of ERAP1 in the ER of JEG-3 cells. A, Subcellular localization of ERAP1 on immunoblot analysis. JEG-3 cell homogenate (T) was fractionated by centrifugation at 10,000 × g for 5 min; the microsomal [pellet (P)] and cytosolic [supernatant (S)] fraction were isolated. B–D, Immunofluorescence staining of JEG-3 cells is shown by confocal laser microscopy for ERAP1 in red (B) and for KDEL, a specific ER marker, in green (C). An overlay of the red and green is shown on the right in yellow (D). The data shown are representative of three independent experiments. Fig. 3 Open in new tabDownload slide Localization of ERAP1 in the ER of JEG-3 cells. A, Subcellular localization of ERAP1 on immunoblot analysis. JEG-3 cell homogenate (T) was fractionated by centrifugation at 10,000 × g for 5 min; the microsomal [pellet (P)] and cytosolic [supernatant (S)] fraction were isolated. B–D, Immunofluorescence staining of JEG-3 cells is shown by confocal laser microscopy for ERAP1 in red (B) and for KDEL, a specific ER marker, in green (C). An overlay of the red and green is shown on the right in yellow (D). The data shown are representative of three independent experiments. Elimination of ERAP1 has no effect on cell surface HLA-G expression in JEG-3 cells under basal conditions We then examined the effect of RNA interference specifically eliminating ERAP1 from JEG-3 cells on cell surface mHLA-G1 expression by flow cytometry to investigate the involvement of ERAP1 in overall antigen presentation to HLA-G. It is noteworthy that newly synthesized mHLA-G1 is unstable and retained in the ER; once mHLA-G1 binds to antigenic peptides in the ER, the stable complexes are transported to the cell surface (29–31). The cell surface expression of HLA-G therefore depends on the supply of antigenic peptides in the ER (14, 15). We obtained remarkably efficient knockdown of ERAP1 in ERAP1-siRNA-treated JEG-3 cells, whereas control siRNA showed no effect on ERAP1 levels by immunoblotting (Fig. 4A). The finding of no apparent changes in the expression of another ER protein, Grp94, in ERAP1 siRNA-treated cells (Fig. 4A) indicated that siRNA treatment caused selective reduction of ERAP1 protein. However, contrary to our expectation that elimination of ERAP1 would reduce cell surface mHLA-G1 levels, ERAP1 siRNA treatment did not decrease them (Fig. 4B). Fig. 4 Open in new tabDownload slide Effect of elimination of ERAP1 on cell surface mHLA-G1 expression in JEG-3 cells. JEG-3 cells were transfected with ERAP1 or control siRNA for 1 d. A, Microsomal fractions from JEG-3 cells treated with siRNA were separated by SDS-PAGE and probed with anti-ERAP1 or anti-Grp94 antibodies. B, Cells were stained with anti-HLA-G1 (MEM-G/9) or irrelevant antibody (background). The mean fluorescence intensities were: background, 4.55 (shaded histogram); ERAP1 siRNA, 177.64 (thin line); and control siRNA, 127.49 (thick line). The data are representative of at least three independent experiments. Fig. 4 Open in new tabDownload slide Effect of elimination of ERAP1 on cell surface mHLA-G1 expression in JEG-3 cells. JEG-3 cells were transfected with ERAP1 or control siRNA for 1 d. A, Microsomal fractions from JEG-3 cells treated with siRNA were separated by SDS-PAGE and probed with anti-ERAP1 or anti-Grp94 antibodies. B, Cells were stained with anti-HLA-G1 (MEM-G/9) or irrelevant antibody (background). The mean fluorescence intensities were: background, 4.55 (shaded histogram); ERAP1 siRNA, 177.64 (thin line); and control siRNA, 127.49 (thick line). The data are representative of at least three independent experiments. LIF induces cell surface mHLA-G1 and ERAP1 expression We failed to show the involvement of ERAP1 in presenting antigenic peptides to HLA-G1 under basal conditions, which led us to infer that ERAP1 is associated with the antigen supply in trophoblastic cells exposed to some stimulants. We decided to specifically examine LIF. JEG-3 cells reportedly have functional LIF receptors (22). At first, we examined whether LIF affects the expression of components of antigen presentation pathway in JEG-3 cells. Flow cytometric analysis showed that LIF increased cell surface expression of mHLA-G1 significantly (mean fluorescence intensity, 65.2 ± 7.8 vs. 281.5 ± 29.3; P < 0.05; Fig. 5). We then examined whether LIF increases ERAP1 levels to evaluate the possible involvement of ERAP1 in this induction. Immunoblotting revealed that LIF increased ERAP1 expression significantly in the microsomal fraction in a time-dependent (Fig. 6, A and B) and concentration-dependent (Fig. 6, C and D) manner, whereas LIF did not change Grp94 expression (Fig. 6, A and C). Moreover, to examine whether putative STAT-3-binding sites in the ERAP1 promoter region mediate this effect of LIF, luciferase constructs containing progressive 5′-deletion fragments of the ERAP1 promoter were transiently transfected into JEG-3 cells (Fig. 6E). LIF stimulated the luciferase activity of the full-length ERAP1 promoter (−1305/+5), which includes three STAT-3-binding sites, by 2.2-fold. Deletions to nucleotide −944 did not affect the magnitude of LIF induction, and deletions to −248 and −176, which resulted in eliminating two upstream STAT-3-binding sites, slightly attenuated the stimulatory effect of LIF to 1.8-fold. However, additional deletion to nucleotide −68, which removed all three STAT-3-binding sites, abrogated the effect. Fig. 5 Open in new tabDownload slide Effect of LIF on cell surface mHLA-G1 expression in JEG-3 cells. JEG-3 cell surface expression of HLA-G1 in the absence (A) or presence (B) of LIF for 48 h was determined using flow cytometry. Cells were stained with anti-HLA-G1 (MEM-G/9) or irrelevant antibody (background). The mean fluorescence intensities were: background, 3.87 (shaded histogram; A and B); HLA-G1 expression without treatment, 185.68 (A; thick line); and after 1000 IU/ml LIF treatment for 48 h, 299.5 (B; thick line). Data are representative of at least three independent experiments. Fig. 5 Open in new tabDownload slide Effect of LIF on cell surface mHLA-G1 expression in JEG-3 cells. JEG-3 cell surface expression of HLA-G1 in the absence (A) or presence (B) of LIF for 48 h was determined using flow cytometry. Cells were stained with anti-HLA-G1 (MEM-G/9) or irrelevant antibody (background). The mean fluorescence intensities were: background, 3.87 (shaded histogram; A and B); HLA-G1 expression without treatment, 185.68 (A; thick line); and after 1000 IU/ml LIF treatment for 48 h, 299.5 (B; thick line). Data are representative of at least three independent experiments. Fig. 6 Open in new tabDownload slide Effect of LIF on ERAP1 expression in JEG-3 cells. A, JEG-3 cells were incubated with 1000 IU/ml LIF, and the time-dependent effect of LIF on ERAP1 expression was examined. The data shown are representative of three separate experiments with independent triplicate samples. B, Quantitative densitometric analyses of ERAP1 proteins normalized for Grp94 proteins (mean ± sd). *, Comparison with the untreated condition is significant (P < 0.05). C, The dose-dependent effect of LIF on ERAP1 expression with of triplicate samples. D, Quantitative densitometric analyses of ERAP1 proteins normalized for Grp94 proteins (mean ± sd). *, Comparison with the untreated condition is significant (P < 0.05). E, After transient transfection, the luciferase activity of each construct was measured in untreated JEG-3 cells (□) and cells treated with 1000 IU/ml LIF for 24 h (▪). The squares indicate the STAT-3-binding sites, which are considered to be LIF-responsive elements. Relative luciferase activity was normalized to the activity of pGL3-basic alone in untreated cells, which was defined as 1.0. These data represent the mean ± sd of triplicate samples. *, Comparison with the untreated condition is significant (P < 0.05). This experiment was carried out in triplicate with similar results. Fig. 6 Open in new tabDownload slide Effect of LIF on ERAP1 expression in JEG-3 cells. A, JEG-3 cells were incubated with 1000 IU/ml LIF, and the time-dependent effect of LIF on ERAP1 expression was examined. The data shown are representative of three separate experiments with independent triplicate samples. B, Quantitative densitometric analyses of ERAP1 proteins normalized for Grp94 proteins (mean ± sd). *, Comparison with the untreated condition is significant (P < 0.05). C, The dose-dependent effect of LIF on ERAP1 expression with of triplicate samples. D, Quantitative densitometric analyses of ERAP1 proteins normalized for Grp94 proteins (mean ± sd). *, Comparison with the untreated condition is significant (P < 0.05). E, After transient transfection, the luciferase activity of each construct was measured in untreated JEG-3 cells (□) and cells treated with 1000 IU/ml LIF for 24 h (▪). The squares indicate the STAT-3-binding sites, which are considered to be LIF-responsive elements. Relative luciferase activity was normalized to the activity of pGL3-basic alone in untreated cells, which was defined as 1.0. These data represent the mean ± sd of triplicate samples. *, Comparison with the untreated condition is significant (P < 0.05). This experiment was carried out in triplicate with similar results. Furthermore, to investigate the possible effects of LIF on other components associated with antigen presentation, we examined the changes in TAP-1, the transporter of antigen precursors to ER. Western blotting showed that LIF at 100 and 1000 IU/ml significantly increased TAP-1 in JEG-3 cells (Fig. 7; P < 0.05). Fig. 7 Open in new tabDownload slide Effect of LIF on TAP expression in JEG-3 cells. A, Expression of TAP1 protein in JEG-3 cells untreated or treated with LIF (48 h; 100 and 1000 IU/ml). The data shown are representative of three separate experiments with independent triplicate samples. B, Quantitative densitometric analyses of TAP1 proteins normalized for β-actin proteins (mean ± sd). *, Comparison with the untreated condition is significant (P < 0.05). Fig. 7 Open in new tabDownload slide Effect of LIF on TAP expression in JEG-3 cells. A, Expression of TAP1 protein in JEG-3 cells untreated or treated with LIF (48 h; 100 and 1000 IU/ml). The data shown are representative of three separate experiments with independent triplicate samples. B, Quantitative densitometric analyses of TAP1 proteins normalized for β-actin proteins (mean ± sd). *, Comparison with the untreated condition is significant (P < 0.05). ERAP1 mediates enhancement of surface expression of HLA-G in LIF-treated cells We observed that LIF induced both cell surface HLA-G and ERAP1 expression, which prompted us to investigate the relationship between HLA-G and ERAP1 in JEG-3 cells under LIF-stimulated conditions. We examined whether loss of ERAP1 might affect the cell surface expression of mHLA-G1 in LIF-treated JEG-3 cells. After LIF treatment, we confirmed that ERAP1-siRNA remarkably reduced ERAP1 proteins by at least 90%, whereas control siRNA did not (Fig. 8A). Flow cytometric analyses demonstrated that when eliminating ERAP1 from LIF-treated cells, the surface expression of mHLA-G1 was suppressed compared with that of cells transfected by control siRNA (Fig. 8B). To confirm this finding in another way, we assessed by immunoblot analysis the release of sHLA-G1, the soluble form of HLA-G1, in the medium of LIF-treated cells (Fig. 8C). In cells treated with control siRNA, LIF increased the secretion of sHLA-G1, whereas elimination of ERAP1 completely reduced the induction to the original levels. ERAP1 siRNA transfection did not change the secretion of sHLA-G in the absence of LIF stimulation. Fig. 8 Open in new tabDownload slide Eliminating ERAP1 affects the cell surface expression of mHLA-G1 and the secretion of sHLA-G in LIF-treated cells. JEG-3 cells were transfected with ERAP1 or control siRNA for 1 d. Cells were then incubated with LIF (1000 IU/ml) or normal medium for another 48 h. A, Microsomal fractions from JEG-3 cells treated with siRNA and LIF (1000 IU/ml) were separated by SDS-PAGE and probed with anti-ERAP1 or anti-Grp94 antibodies. B, Cells were stained with anti-HLA-G1 (MEM-G/9) or irrelevant antibody (background). The mean fluorescence intensities were: background, 4.55 (shaded histogram); ERAP1 siRNA, 94.39 (thin line); and control siRNA, 168.87 (thick line). C, Culture supernatant fractions from siRNA- and LIF-treated JEG-3 cells were separated by SDS-PAGE and probed with anti-HLA-G1 antibody (16G1), which reacts only with sHLA-G. The data shown are representative of three separate experiments with independent triplicate samples (upper panel). Quantitative densitometric analyses of sHLA-G are shown (mean ± sd; lower panel). *, Comparison is significant (P < 0.05). Fig. 8 Open in new tabDownload slide Eliminating ERAP1 affects the cell surface expression of mHLA-G1 and the secretion of sHLA-G in LIF-treated cells. JEG-3 cells were transfected with ERAP1 or control siRNA for 1 d. Cells were then incubated with LIF (1000 IU/ml) or normal medium for another 48 h. A, Microsomal fractions from JEG-3 cells treated with siRNA and LIF (1000 IU/ml) were separated by SDS-PAGE and probed with anti-ERAP1 or anti-Grp94 antibodies. B, Cells were stained with anti-HLA-G1 (MEM-G/9) or irrelevant antibody (background). The mean fluorescence intensities were: background, 4.55 (shaded histogram); ERAP1 siRNA, 94.39 (thin line); and control siRNA, 168.87 (thick line). C, Culture supernatant fractions from siRNA- and LIF-treated JEG-3 cells were separated by SDS-PAGE and probed with anti-HLA-G1 antibody (16G1), which reacts only with sHLA-G. The data shown are representative of three separate experiments with independent triplicate samples (upper panel). Quantitative densitometric analyses of sHLA-G are shown (mean ± sd; lower panel). *, Comparison is significant (P < 0.05). Discussion The EVTs express nonclassical HLA class I molecules such as HLA-G, which are likely to play salient roles in protecting EVTs from NK cell-mediated lysis at the time of implantation. However, regarding the immune system, the manner in which antigenic peptides are produced in the ER to interact with HLA-G remains to be determined. This study demonstrated that ERAP1 was located in the ER of trophoblastic cells and enhanced cell surface HLA-G1 expression under LIF stimulation, which signified the increased supply of antigenic peptides in the ER. This is the first report elucidating the functional role of ERAP1 in trophoblastic cells; ERAP1 would be involved in antigen presentation of nonclassical HLA class I molecules. Our finding that ERAP1 protein was expressed specifically in EVTs from explant cultures of human chorionic villi is consistent with previous results obtained through immunohistochemical analyses of human placental sections (17). Because it would be better to use these normal EVTs to study the function of ERAP1 in the placenta, we initially attempted to use them in our study, but failed to obtain efficient transfection and reproducible data. We therefore used choriocarcinoma cells, BeWo and JEG-3, which are malignant neoplasms that represent early trophoblasts at implantation or later invasive stages and serve as a valid in vitro model for studying the function of trophoblasts (25, 26). We observed predominant ERAP1 expression in JEG-3 more than in BeWo cells, similar to the report of another aminopeptidase, dipeptidyl peptidase IV, showing abundant expression in JEG-3 and negligible expression in BeWo cells (28). Although JEG-3 cells were derived originally from BeWo cells, BeWo cells might resemble normal villous trophoblasts more closely than JEG-3. That inference is supported by the finding that differentiation and invasion tend to be inversely correlated in normal trophoblasts and BeWo cells, but not in JEG-3 cells (32). Our finding that HLA-G was expressed in JEG-3 cells more abundantly than in BeWo cells might also corroborate the validity of JEG-3 cells as a suitable model for EVTs. Efforts to clarify how MHC class I-presented peptides are generated have recently emphasized the requirement for several proteolytic steps of precursor proteins (9–11). After rough degradation of precursor proteins by proteasomes, aminopeptidases in the ER trimmed oligopeptides into peptides of the correct size. Peptides longer or shorter than eight or nine residues bind weakly to MHC class I, which are further degraded and become unable to function as antigen. ERAP1 is the first to be identified as this trimming enzyme in somatic cells (14, 15). Although the intracellular location of ERAP1 in trophoblastic cells remained unclear, we were able to demonstrate that ERAP1 was present together with the ER marker KDEL sequence in JEG-3 cells. Because the ERAP1 gene does not contain the KEDL sequence (12), colocalization of ERAP1 with KDEL strongly indicated the location of ERAP1 in the ER, suggesting the possible involvement of ERAP1 in trimming antigenic peptides in trophoblasts also. To investigate the antigen supply by ERAP1 in JEG-3 cells, we monitored the amount of cell surface mHLA-G1 expression, which reflects the supply of antigenic peptides in the ER (14, 15). However, contrary to our expectation, eliminating ERAP1 did not reduce cell surface mHLA-G1 expression in JEG-3 cells under basal conditions. This finding was similar to that in HeLa cells; eliminating ERAP1 from HeLa cells with no additives does not decrease overall cell surface expression of MHC class I molecules (15). This observation has been rationalized by the possibility that ERAP1 might reduce surface amounts of peptide-classical MHC class I complexes by destroying nine-residue antigenic peptides under basal conditions (15). HLA-G predominantly binds to peptides of nine amino acids, especially those that contain sequences with leucine at the C terminus and proline at position 3 (27). However, this type of nine-residue peptide, such as FAPGNYPAL, might also be susceptible to ERAP1, leading to a decrease in antigenic peptides. Once antigen-MHC complexes are generated, ERAP1 can no longer attack the peptides. Although speculative, under basal conditions, HLA-G itself might be insufficient to form stable antigen-MHC complexes, which results in more antigen-destroying activity than antigen trimming by ERAP1. As another possibility, some other ER aminopeptidases might compensate for the knockdown of ERAP1 in trimming antigenic peptides, such as ERAP2. Reportedly, ERAP2 is the dominant trimming enzyme in tissues with low ERAP1 expression (33), which resembles the effect of ERAP1 siRNA treatment. Additional studies to examine the expression of ERAP2 might clarify this possibility. As factors required for maternal immune tolerance, we focused our attention on LIF, because it is produced abundantly from the endometrium during the implantation period of normal pregnancy, and production of LIF in women with unexplained recurrent abortions is decreased (34, 35). Our finding of ERAP1 induction by LIF in JEG-3 cells was the first to show the effect of LIF on ERAP1 levels in any type of cell. We also confirmed up-regulation of ERAP1 at transcriptional levels, which was suggested by promoter analyses (23). An approximately 900-bp region from −944 to −68 containing three STAT-3-binding sites was required for mediating the maximal stimulatory effect of LIF on promoter activity. In contrast to the slight attenuation that was achieved by deleting two upstream STAT-3-binding sites, the deletion of the proximal site completely abrogated the LIF stimulatory effects, implying the essential contribution of that proximal site to LIF stimulation. Additional experiments introducing point mutations to the STAT-3-binding sites will clarify this possibility. Reportedly, LIF increases HLA-G promoter activity in JEG-3 cells (22). In addition to this direct effect, we speculated that LIF increases the cell surface expression of mHLA-G1 at least in part via up-regulating ERAP1. As expected, eliminating ERAP1 reduced the cell surface expression of HLA-G after LIF-treatment in JEG-3 cells, which strongly suggests that ERAP1 is associated with antigen presentation to HLA-G in trophoblasts under some physiological conditions. The relation between LIF and HLA-G1 in JEG-3 cells is similar to the finding that eliminating ERAP1 reduces surface MHC class I expression after IFN-γ treatment in HeLa cells (14). During the implantation period, with abundant LIF production, ERAP1 would increase the supply of antigenic peptides to HLA-G via trimming N-extended antigenic precursors, which functions to establish maternal immune tolerance concomitantly with the up-regulation of TAP and HLA-G. Although we used JEG-3 cells as a model for normal trophoblasts, our data could also be interpreted from the viewpoint of choriocarcinoma development. Gestational choriocarcinomas occur after normal or abnormal pregnancy, including hydatidiform mole. Choriocarcinomas, therefore, are unique in that they express paternal genes and are an allograft in the maternal host. Like normal trophoblasts, cell surface expression of HLA-G in choriocarcinomas would be involved in preventing the recognition by NK cells (36). Interestingly, LIF induces the proliferation and invasion of choriocarcinoma cells (37). In addition to this direct influence, LIF may function to facilitate the growth of choriocarcinomas through promoting the generation of HLA-G-antigenic peptide complexes in patients with choriocarcinomas, given that LIF levels increase where choriocarcinomas exist. In conclusion, this study indicates that in the presence of LIF, ERAP1 plays a key role in presenting antigenic peptides to HLA-G in the ER in JEG-3 choriocarcinoma trophoblastic cells. Our findings may provide a clue to understanding the molecular mechanisms of immune tolerance at the feto-maternal interface. Acknowledgments We are grateful to Dr. Daniel. E. Geraghty (Fred Hutchinson Cancer Research Center, Seattle, WA) for generously providing the anti-HLA-G antibody (16G1). This work was supported in part by grants-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and from the Ministry of Internal Affairs and Communications of Japan for specific medical research (in collaboration with Nagoya Teishin Hospital). F.S., T.I., S.N., E.Y., S.S., K.I., A.I., A.H., M.T., S.M., and F.K. have nothing to declare. Abbreviations Ab Antibody A-LAP adipocyte-derived leucine aminopeptidase ER endoplasmic reticulum ERAP1 endoplasmic reticulum aminopeptidase-1 EVT extravillous trophoblast HLA human leukocyte antigen IFN-γ interferon-γ KDEL Lys-Asp-Glu-Leu LIF leukemia inhibitory factor mAb monoclonal antibody mHLA membrane-bound human leukocyte antigen MHC major histocompatibility complex NK natural killer sHLA soluble isoform of human leukocyte antigen siRNA small interfering RNA STAT-3 signal transducer and activator of transcription-3 TAP transporter associated with antigen processing 1 Loke YW , Hiby S , King A 1999 Human leucocyte antigen-G and reproduction. J Reprod Immunol 43 : 235 – 242 Google Scholar Crossref Search ADS PubMed WorldCat 2 Faulk WP , Temple A 1976 Distribution of β2 microglobulin and HLA in chorionic villi of human placentae. Nature 262 : 799 – 802 Google Scholar Crossref Search ADS PubMed WorldCat 3 Kovats S , Main EK , Librach C , Stubblebine M , Fisher SJ , DeMars R 1990 A class I antigen, HLA-G, expressed in human trophoblasts. Science 248 : 220 – 223 Google Scholar Crossref Search ADS PubMed WorldCat 4 Lanier LL 1999 Natural killer cells fertile with receptors for HLA-G? Proc Natl Acad Sci USA 96 : 5343 – 5345 Google Scholar Crossref Search ADS PubMed WorldCat 5 Ishitani A , Geraghty DE 1992 Alternative splicing of HLA-G transcripts yields proteins with primary structures resembling both class I and class II antigens. Proc Natl Acad Sci USA 89 : 3947 – 3951 Google Scholar Crossref Search ADS PubMed WorldCat 6 Park B , Ahn K 2003 An essential function of tapasin in quality control of HLA-G molecules. J Biol Chem 278 : 14337 – 14345 Google Scholar Crossref Search ADS PubMed WorldCat 7 Munz C , Nickolaus P , Lammert E , Pascolo S , Stevanovic S , Rammensee HG 1999 The role of peptide presentation in the physiological function of HLA-G. Semin Cancer Biol 9 : 47 – 54 Google Scholar Crossref Search ADS PubMed WorldCat 8 Mallet V , Proll J , Solier C , Aguerre-Girr M , DeRossi M , Loke YW , Lenfant F , Le Bouteiller P 2000 The full length HLA-G1 and no other alternative form of HLA-G is expressed at the cell surface of transfected cells. Hum Immunol 61 : 212 – 224 Google Scholar Crossref Search ADS PubMed WorldCat 9 Mo XY , Cascio P , Lemerise K , Goldberg AL , Rock K 1999 Distinct proteolytic processes generate the C and N termini of MHC class I-binding peptides. J Immunol 163 : 5851 – 5859 Google Scholar PubMed WorldCat 10 Brouwenstijn N , Serwold T , Shastri N 2001 MHC class I molecules can direct proteolytic cleavage of antigenic precursors in the endoplasmic reticulum. Immunity 15 : 95 – 104 Google Scholar Crossref Search ADS PubMed WorldCat 11 Goldberg AL , Cascio P , Saric T , Rock KL 2002 The importance of the proteasome and subsequent proteolytic steps in the generation of antigenic peptides. Mol Immunol 39 : 147 – 164 Google Scholar Crossref Search ADS PubMed WorldCat 12 Hattori A , Matsumoto H , Mizutani S , Tsujimoto M 1999 Molecular cloning of adipocyte-derived leucine aminopeptidase highly related to placental leucine aminopeptidase/oxytocinase. J Biochem 125 : 931 – 938 Google Scholar Crossref Search ADS PubMed WorldCat 13 Hattori A , Kitatani K , Matsumoto H , Miyazawa S , Rogi T , Tsuruoka N , Mizutani S , Natori Y , Tsujimoto M 2000 Characterization of recombinant human adipocyte-derived leucine aminopeptidase expressed in Chinese hamster ovary cells. J Biochem 128 : 755 – 762 Google Scholar Crossref Search ADS PubMed WorldCat 14 Saric T , Chang SC , Hattori A , York IA , Markant S , Rock KL , Tsujimoto M , Goldberg AL 2002 An IFN-γ-induced aminopeptidase in the ER, ERAP1, trims precursors to MHC class I-presented peptides. Nat Immunol 3 : 1169 – 1176 Google Scholar Crossref Search ADS PubMed WorldCat 15 York IA , Chang SC , Saric T , Keys JA , Favreau JM , Goldberg AL , Rock KL 2002 The ER aminopeptidase ERAP1 enhances or limits antigen presentation by trimming epitopes to 8–9 residues. Nat Immunol 3 : 1177 – 1184 Google Scholar Crossref Search ADS PubMed WorldCat 16 Serwold T , Gonzalez F , Kim J , Jacob R , Shastri N 2002 ERAAP customizes peptides for MHC class I molecules in the endoplasmic reticulum. Nature 419 : 480 – 483 Google Scholar Crossref Search ADS PubMed WorldCat 17 Ino K , Kikkawa F , Suzuki T , Kajiyama H , Shibata K , Nomura S , Itakura A , Ito M , Nagasaka T , Hattori A , Tsujimoto M , Mizutani S 2003 Expression of placental leucine aminopeptidase and adipocyte-derived leucine aminopeptidase in human normal and malignant invasive trophoblastic cells. Lab Invest 83 : 1799 – 1809 Google Scholar Crossref Search ADS PubMed WorldCat 18 Norwitz ER , Schust DJ , Fisher SJ 2001 Implantation and the survival of early pregnancy. N Engl J Med 345 : 1400 – 1408 Google Scholar Crossref Search ADS PubMed WorldCat 19 Stewart CL , Kaspar P , Brunet LJ , Bhatt H , Gadi I , Kontgen F , Abbondanzo SJ 1992 Blastocyst implantation depends on maternal expression of leukaemia inhibitory factor. Nature 359 : 76 – 79 Google Scholar Crossref Search ADS PubMed WorldCat 20 Nachtigall MJ , Kliman HJ , Feinberg RF , Olive DL , Engin O , Arici A 1996 The effect of leukemia inhibitory factor (LIF) on trophoblast differentiation: a potential role in human implantation. J Clin Endocrinol Metab 81 : 801 – 806 Google Scholar PubMed WorldCat 21 Sharkey AM , King A , Clark DE , Burrows TD , Jokhi PP , Charnock-Jones DS , Loke YW , Smith SK 1999 Localization of leukemia inhibitory factor and its receptor in human placenta throughout pregnancy. Biol Reprod 60 : 355 – 364 Google Scholar Crossref Search ADS PubMed WorldCat 22 Bamberger AM , Jenatschke S , Schulte HM , Loning T , Bamberger MC 2000 Leukemia inhibitory factor (LIF) stimulates the human HLA-G promoter in JEG3 choriocarcinoma cells. J Clin Endocrinol Metab 85 : 3932 – 3936 Google Scholar Crossref Search ADS PubMed WorldCat 23 Hattori A , Matsumoto H , Mizutani S , Tsujimoto M 2001 Genomic organization of the human adipocyte-derived leucine aminopeptidase gene and its relationship to the placental leucine aminopeptidase/oxytocinase gene. J Biochem 130 : 235 – 241 Google Scholar Crossref Search ADS PubMed WorldCat 24 Bousquet C , Zatelli MC , Melmed S 2000 Direct regulation of pituitary proopiomelanocortin by STAT3 provides a novel mechanism for immuno-neuroendocrine interfacing. J Clin Invest 106 : 1417 – 1425 Google Scholar Crossref Search ADS PubMed WorldCat 25 Ringler GE , Strauss III JF 1990 In vitro systems for the study of human placental endocrine function. Endocr Rev 11 : 105 – 123 Google Scholar Crossref Search ADS PubMed WorldCat 26 Avril T , Jarousseau AC , Watier H , Boucraut J , Le Bouteiller P , Bardos P , Thibault G 1999 Trophoblast cell line resistance to NK lysis mainly involves an HLA class I-independent mechanism. J Immunol 162 : 5902 – 5909 Google Scholar PubMed WorldCat 27 Lee N , Malacko AR , Ishitani A , Chen MC , Bajorath J , Marquardt H , Geraghty DE 1995 The membrane-bound and soluble forms of HLA-G bind identical sets of endogenous peptides but differ with respect to TAP association. Immunity 3 : 591 – 600 Google Scholar Crossref Search ADS PubMed WorldCat 28 Sato Y , Fujiwara H , Higuchi T , Yoshioka S , Tatsumi K , Maeda M , Fujii S 2002 Involvement of dipeptidyl peptidase IV in extravillous trophoblast invasion and differentiation. J Clin Endocrinol Metab 87 : 4287 – 4296 Google Scholar Crossref Search ADS PubMed WorldCat 29 York IA , Goldberg AL , Mo XY , Rock KL 1999 Proteolysis and class I major histocompatibility complex antigen presentation. Immunol Rev 172 : 49 – 66 Google Scholar Crossref Search ADS PubMed WorldCat 30 Shastri N , Schwab S , Serwold T 2002 Producing nature’s gene-chips: the generation of peptides for display by MHC class I molecules. Annu Rev Immunol 20 : 463 – 493 Google Scholar Crossref Search ADS PubMed WorldCat 31 Rock KL , York IA , Saric T , Goldberg AL 2002 Protein degradation and the generation of MHC class I-presented peptides. Adv Immunol 80 : 1 – 70 Google Scholar PubMed WorldCat 32 Hohn HP , Linke M , Ugele B , Denker HW 1998 Differentiation markers and invasiveness: discordant regulation in normal trophoblast and choriocarcinoma cells. Exp Cell Res 244 : 249 – 258 Google Scholar Crossref Search ADS PubMed WorldCat 33 Tanioka T , Hattori A , Masuda S , Nomura Y , Nakayama H , Mizutani S , Tsujimoto M 2003 Human leukocyte-derived arginine aminopeptidase. The third member of the oxytocinase subfamily of aminopeptidases. J Biol Chem 278 : 32275 – 32283 Google Scholar Crossref Search ADS PubMed WorldCat 34 Piccinni MP , Beloni L , Livi C , Maggi E , Scarselli G , Romagnani S 1998 Defective production of both leukemia inhibitory factor and type 2 T-helper cytokines by decidual T cells in unexplained recurrent abortions. Nat Med 4 : 1020 – 1024 Google Scholar Crossref Search ADS PubMed WorldCat 35 Steck T , Giess R , Suetterlin MW , Bolland M , Wiest S , Poehls UG , Dietl J 2004 Leukaemia inhibitory factor (LIF) gene mutations in women with unexplained infertility and recurrent failure of implantation after IVF and embryo transfer. Eur J Obstet Gynecol Reprod Biol 112 : 69 – 73 Google Scholar Crossref Search ADS PubMed WorldCat 36 Saji F , Kameda T , Koyama M , Matsuzaki N , Negoro T , Tanozawa O 1989 Impaired susceptibility of human trophoblast to MHC nonrestricted killer cells: implication in the maternal-fetal relationship. Am J Reprod Immunol 19 : 108 – 113 Google Scholar Crossref Search ADS PubMed WorldCat 37 Fitzgerald JS , Tsareva SA , Poehlmann TG , Berod L , Meissner A , Corvinus FM , Wiederanders B , Pfitzner E , Markert UR , Friedrich K 2005 Leukemia inhibitory factor triggers activation of signal transducer and activator of transcription 3, proliferation, invasiveness, and altered protease expression in choriocarcinoma cells. Int J Biochem Cell Biol 37 : 2284 – 2296 Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2006 by The Endocrine Society
TI - Endoplasmic Reticulum Aminopeptidase-1 Mediates Leukemia Inhibitory Factor-Induced Cell Surface Human Leukocyte Antigen-G Expression in JEG-3 Choriocarcinoma Cells
JO - Endocrinology
DO - 10.1210/en.2005-1449
DA - 2006-04-01
UR - https://www.deepdyve.com/lp/oxford-university-press/endoplasmic-reticulum-aminopeptidase-1-mediates-leukemia-inhibitory-dKZsOvJmgr
SP - 1780
VL - 147
IS - 4
DP - DeepDyve
ER -