HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Morales, P. J. Articles by Hunt, J. S. Search for Related Content PubMed PubMed Citation Articles by Morales, P. J. Articles by Hunt, J. S.

Placental Cell Expression of HLA-G2 Isoforms Is Limited to the Invasive Trophoblast Phenotype ¹

Pedro J. Morales^*, Judith L. Pace, Jeralyn Sue Platt^*, Teresa A. Phillips^*, Kim Morgan^2,*, Asgi T. Fazleabas and Joan S. Hunt^3,*,

^* Departments of Anatomy and Cell Biology, Molecular and Integrative Physiology, and Pathology and Laboratory Medicine, University of Kansas Medical Center, Kansas City, KS 66160; and Department of Obstetrics and Gynecology, University of Illinois, Chicago, IL 60612

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The HLA-G message is alternatively spliced into multiple transcripts, two of which encode soluble isoforms. To initiate studies on the specific functions of the soluble isoforms, we produced soluble rHLA-G1 (rsG1) and rsG2 in human embryonic kidney 293 cells and characterized the proteins. Both isoforms were glycosylated and formed disulfide-bonded oligomers. Recombinant sG1 associated with ₂-microglobulin, whereas rsG2 did not. Mouse mAb generated to rsG1 (1-2C3), which identified exclusively sG1, and mAb generated to rsG2 (26-2H11), which identified both soluble and membrane G2 (m/sG2), were used for immunohistochemical isoform mapping studies on placental tissue sections. Soluble G1 protein was abundant in many subpopulations of trophoblast cells, whereas m/sG2 protein was present exclusively in extravillous cytotrophoblast cells. Although both isolated placental villous cytotrophoblast cells and chorion membrane extravillous cytotrophoblast cells contained mRNAs encoding sG1 and sG2, protein expression was as predicted from the immunostains with m/sG2 present only in the invasive trophoblast subpopulation. Analysis of function by Northern and Western blotting demonstrated that both rsG1 and rsG2 inhibit CD8 expression on PBMC without changing CD3 expression or causing apoptotic cell death. Collectively, the studies indicate that: 1) both sG1 and m/sG2 are produced in placentas; 2) transcription and translation are linked for sG1, but not G2; 3) expression of G2 is exclusively associated with the invasive phenotype; and 4) the two isoforms of sG may promote semiallogeneic pregnancy by reducing expression of CD8, a molecule required for functional activation of CTL.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Considerable information has been acquired on the class Ib HLA-G gene first identified by Geraghty et al. in 1987 (1). HLA-G is characterized by low polymorphism, a 16-bp deletion within the enhancer A/IFN consensus sequence, and a nonfunctional IFN--activated site (GAS)⁴ sequence (2, 3, 4, 5, 6). Multiple transcripts encoding both membrane-bound and soluble isoforms are generated by alternative splicing of the mRNA (7). Less is known of the proteins encoded by the transcripts, although it has been determined that at least three of the isoforms of HLA-G are membrane bound; HLA-G1 is the full-length isoform, whereas HLA-G2 and HLA-G3 lack the 2 and 3 domains, respectively. The two soluble isoforms, soluble HLA-G1 (sG1) and sG2, also known as HLA-G5 and HLA-G6, appear to resemble membrane HLA-G1 and HLA-G2, except for being truncated by a stop codon in the intron 4 region, which prevents translation of the transmembrane region (8). The secondary and tertiary characteristics of the proteins remain poorly understood, although it has been proposed that HLA-G2 proteins form homodimers and have HLA class II-like rather than class I-like characteristics (7).

HLA-G is reportedly characterized by a degree of tissue-specific expression, with the placenta identified as a major site of production of HLA-G messages and proteins (9, 10, 11). In the placental bed, where genetically different maternal and fetal cells reside in apparent accord, the unusual characteristics of the HLA-G gene may be of importance. For example, low polymorphism could ensure that most mothers will fail to recognize trophoblast-presented HLA-G as foreign (12). The deletion in the promoter region and substitution in the GAS element (1, 5), which preclude vigorous responses to endogenous and exogenous IFNs, could stabilize HLA-G expression through implantation and parturition. The spectrum of messages encoding different isoforms could encode proteins with distinctly different structures and functions, which would perhaps permit the polypeptides to perform different functions, as is the case with CD45 (13).

Many studies have now addressed functional aspects of HLA-G1 (14, 15, 16), but none has explored HLA-G2, the soluble isoform that appears to circulate in maternal blood throughout pregnancy (17) and is believed to compensate when HLA-G1 is absent (18). The studies reported in this work were therefore designed to investigate and compare structural aspects of sG1 and sG2 and to initiate functional experiments by determining: 1) whether either isoform is positioned appropriately at the maternal-fetal interface for influencing maternal immune cells, and 2) whether either isoform exerts such an influence.

To achieve these goals, we first produced rsG1 and rsG2 in eukaryotic cells so as to compare their biochemical features, then generated mAb to the two recombinant proteins. The new mAb were characterized and used to identify HLA-G soluble isoforms in tissues. In a final group of experiments designed to investigate the functional capacity of the rsG1 and rsG2 proteins, particularly their ability to modulate maternal immune cells, we studied the effects of rsG1 and rsG2 on expression of a major CTL Ag, CD8, in activated PBMC.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Isolation and cloning of HLA-G transcripts

cDNAs encoding HLA-G transcripts were isolated from S14/8 cells, a mouse fibroblast cell line stably transfected with a 6.0-kb HindIII fragment of genomic DNA containing the full-length HLA-G gene (gift from B. Koller, University of North Carolina, Chapel Hill, NC, and H. Orr, University of Minnesota, Minneapolis, MN), JEG-3 (HTB-36), and Jar (HTB-144) choriocarcinoma cells obtained from American Type Culture Collection (ATCC, Manassas, VA), as well as villous trophoblast cells and extravillous cytotrophoblast cells isolated and purified, as described below. Total RNA was prepared using TRIzol (Invitrogen, Carlsbad, CA) and treated with DNase I (Sigma-Aldrich, St. Louis, MO). Transcripts encoding membrane HLA-G1 (mG1) and HLA-G2 (mG2) were isolated from villous and extravillous cytotrophoblast cells, respectively, by RT-PCR using G45–65 (5'-GCC CTG ACC CTG ACC GAG AC-3') as the forward primer and G1225 as the reverse primer (19), then were TA cloned into pGEM-T Easy Vector System I (Promega, Madison, WI). To develop expression constructs for mG1 and mG2, the inserts were released by EcoRI digestion, purified using a QIAquick gel extraction kit (Qiagen, Valencia, CA). Restriction sites for cloning into the modified expression vector pRC/CMV (20) (Invitrogen), a gift from B. Hudson (University of Kansas Medical Center), were added (NheI: A-NheIMTW, 5'-CTA GCT AGC TAG AGG CTC CCA CTC CA-3'; XbaI: B-mXbaIMTW, 5'-CTA GTC TAG ACT AGA TCA ATC TGA GCT CTT-3'). To obtain transcripts encoding sG1 and sG2 from S14/8 cell RNA, two PCR were performed. Messages obtained by RT-PCR using the G45–65 primer and G1225 were purified as above, then subjected to a second PCR using the following cloning primers: A-NheIMTW, 5'-CTA GCT AGC TAG AGG CTC CCA CTC CA-3'; B-XbaIMTW, 5'-CTA GTC TAG ACT AGA TTA AAG GTC TTC AGA-3'. Amplicons containing NheI and XbaI sites were inserted into the NheI and XbaI sites of the modified vector pRC/CMV (20) (Invitrogen). Plasmid constructs were sequenced by using the ABI PRISIM XL BIG DYE sequencing system in the Biotechnology Support Facility at the University of Kansas Medical Center.

Transfections

Human embryonic kidney cells (HEK293, ATCC, CRL-1573) were stably transfected with the constructs encoding sG1 and sG2. HEK293 cells were maintained at 37°C in 5% CO₂ in DMEM supplemented with 10% FBS and penicillin (100 U/ml) with streptomycin (0.1 mg/ml). Transfection was performed using LipofectAMINE Plus (Invitrogen), according to the manufacturer’s specifications. Stable transfectants were selected using G418 (Invitrogen) and cloned by limiting dilution. Selection of rsG1- and rsG2-producing clones (patents pending) was achieved by using an ELISA to test for the proteins. HEK293 cells were transiently transfected to obtain mG1 and mG2 proteins using essentially the same system, and were tested for cell surface expression of mG1 and mG2 by flow cytometry after 24 h in culture.

Preparation of purified rsG1 and rsG2

Recombinant sG1 and rsG2 proteins were purified from supernatant culture medium by immunoaffinity chromatography using an anti-FLAG-M2 affinity resin (Sigma-Aldrich). Each lot was tested for LPS using the Pyrotell detection gel-clot formulation Limulus amebocyte lysate assay (Associates of Cape Cod, Falmouth, MA). Only LPS-free lots were used in functional experiments.

Immunoblotting

Immunoblotting was used to identify rsG1 and rsG2 purified from culture supernatants as well as to determine association of ₂-microglobulin (₂m) with these proteins. The technique was also used to identify cell surface markers on PBMC and to detect cleavage products in an apoptosis assay. In brief, proteins were resuspended in Laemmli buffer (21) under reducing conditions, separated by electrophoresis in acrylamide gels (10 or 15% SDS-PAGE), and electrotransferred onto nitrocellulose membranes (Schleicher & Schuell, Keene, NH). Recombinant sG1 and rsG2 were detected using the mAb 16G1, which is specific for HLA-G intron 4 amino acids (a gift of D. Geraghty, Fred Hutchinson Cancer Research Center, Seattle, WA) (0.2 µg/ml) and anti-FLAG M1 Ab (Sigma-Aldrich) (1 µg/ml). ₂m was identified using a mAb from Amac (Westbrook, ME), at 1 µg/ml. To detect cell surface markers on PBMC, goat anti-human CD8 and mouse anti-human CD3, both from Santa Cruz Biotechnology (Santa Cruz, CA), and rabbit anti-human -actin (Sigma-Aldrich) were used. In these experiments, each blot was hybridized sequentially for CD8, CD3, and -actin. Apoptosis assays were done by investigating poly(ADP-ribose) polymerase (NAD⁺ ADP ribosyltransferase, EC 2.4.2.30) (PARP) cleavage (Oncogene Research Products, Boston, MA). For all immunoblots, signal was detected using an appropriate secondary Ab and SuperSignal West Pico (Pierce, Rockford, IL). Thereafter, the blots were exposed to Hyperfilm ECL (Amersham Pharmacia Biotech, Piscataway, NJ).

Detection of glycosylation

To establish glycosylation patterns, purified rsG1 and rsG2 were digested under native and denaturing conditions using an enzymatic deglycosylation kit that cleaves N-linked and sialic acid-substituted Gal 1–3 GalNAc 1 O-linked oligosaccharides from glycoproteins (Bio-Rad, Hercules, CA). After enzymatic digestion, the samples were denatured in reducing Laemmli buffer, separated on SDS-PAGE, transferred onto nitrocellulose membranes, and detected with the mAb 16G1, as described above.

Detection of oligomerization

Disulfide-based self dimerization was tested by running 4 µg of each purified protein on a 4–20% gradient SDS-PAGE under reducing and nonreducing conditions, transferring and detecting using 16G1, as before.

ELISA

Recombinant protein production was monitored by ELISA, which were constructed and performed essentially as previously described (22), with the exception that the microwells (replicates of three) were coated with supernatant culture medium from the transfected HEK293 clones. The coated plates were incubated with either 16G1, anti-₂m (Amac), W6/32 (ATCC, HB-95), normal mouse IgG (Vector Laboratories, Burlingame, CA), or isotype controls IgG1 and IgG2a (BD PharMingen, San Diego, CA). ELISA were also constructed and used to characterize the new mAbs described below. The new mAbs were tested for reactivity to pooled HLA class I Ags purified from human platelets (a gift from K. Kao, University of Florida, Gainesville, FL) (23), FLAG-purified rsG1 and rsG2, purified ₂m (Sigma-Aldrich), and intron 4 peptide KEGDGGIMSVRESRSLSEDL (Biotechnology Support Facility, University of Kansas Medical Center). For both ELISA, bound Ab was detected by the addition of HRP-conjugated anti-mouse IgG (Vector Laboratories), followed by tetramethylbenzidine substrate (Kirkegaard & Perry, Gaithersburg, MD). After stopping the reaction with 1 M H₃PO₄, the plates were read at 450 nm with an ELx 808 microplate reader (Bio-Tek Instruments, Winnoski, VT).

mAb generation and characterization

mAbs to rsG1 and rsG2 were generated in the facility of the University of Illinois (Urbana-Champaign, IL) using standard techniques (24, 25). Ten-week-old female BALB/c mice were immunized by i.p. injection using 25 µg of rsG1 or rsG2 for each injection. Mouse sera that screened positive for Ab reactive to rsG1 or rsG2 were sacrificed, and standard hybridoma technology was applied. After 2–3 wk, hypoxanthine/aminopterin/thymidine-resistant cultures were isolated, and supernatant culture medium was screened for reactivity to rsG1 and rsG2 by ELISA. Hybridomas producing Ab exclusively reactive with either rsG1 or rsG2 were cloned by limiting dilution, expanded, and frozen.

Flow cytometry assays

Flow cytometry experiments were performed according to standard methods. Briefly, HEK293 cells transiently transfected with vector, mG1, or mG2 were tested using the following mAb: W6/32 (2.5 µg/ml), 1-2C3 (10 µg/ml), 26-2H11 (10 µg/ml), or isotype control (IgG1 or IgG2a, 10 and 2.5 µg/ml, respectively). Untransfected (721.221) and HLA-G1-transfected (721.221-G1) lymphoblastoid cells were tested using anti-HLA-G1 (87G, 5 µg/ml), 1-2C3 (2.5 µg/ml), and isotype controls, as before. Cells were incubated with mAb on ice for 30 min, then were washed four times with buffer (PBS containing 0.1% sodium azide and 1% BSA). Bound Ab was detected by staining with the F(ab')₂ of R-PE-conjugated sheep anti-mouse IgG (Sigma-Aldrich). Cells were fixed in PBS containing 1% paraformaldehyde and analyzed by flow cytometry using a FACSCalibur and CellQuest software (BD Biosciences, San Diego, CA).

Tissue collection and purification of villous and extravillous cytotrophoblast cells

First trimester and term placentas as well as extraplacental membranes were obtained from elective pregnancy terminations and normal term deliveries, respectively, in accordance with a protocol approved by the Human Subjects Committee of the University of Kansas Medical Center. For immunohistochemical experiments, random areas of the tissues were manually dissected into 1-cm² sections, embedded in TBS (tissue-freezing medium; Triangle Biomedical Sciences, Durham, NC), and stored at -80°C until sectioned for immunohistology. Villous cytotrophoblast cells were collected from term placentas using a previously described protocol that included removal of HLA class I-positive cells by magnetic bead technology (26). The cell suspensions were characterized by centrifuging onto glass slides using a Shandon (Pittsburgh, PA) Cytospin and immunostaining to identify trophoblast cells using anti-cytokeratin-7 (DAKO, Carpinteria, CA) and macrophages using anti-CD14 (Zymed Laboratories, South San Francisco, CA). Purity was routinely >95%. Extravillous cytotrophoblast cells were collected from chorion membranes, as follows. Briefly, the chorion membrane was peeled from the amnion membrane and manually scraped free of decidua. The chorion membrane was minced thoroughly with scissors, rinsed to remove blood, and subjected to three enzymatic digests in freshly prepared digestion solution (100 U/ml penicillin, 100 µg/ml streptomycin, 20 mM HEPES, 30 mM sodium bicarbonate, 10 mg/ml BSA, 200 U/ml collagenase, 1 mg/ml hyaluronidase, 150 µg/ml DNase in HBSS) at 37°C for 20 min, followed by gradient centrifugation over Histopaque 1077 (Sigma-Aldrich). The enriched population of extravillous cytotrophoblasts was depleted of CD14⁺ macrophages by negative selection using an MS separation column (Miltenyi Biotec, Auburn, CA) with mouse anti-CD14 Ab (Zymed Laboratories). The collected cells were analyzed in cytospin preparations for the same markers as before, and were identified as >95% cytokeratin-7 positive.

Immunohistochemistry and immunocytochemistry

Two sections of frozen first trimester placenta, term placenta, or term amniochorion membrane, or two spots of cytospun trophoblast cells were placed onto glass slides. The tissues and cells were evaluated by immunohistochemistry using previously described techniques (27). Abs used included anti-cytokeratin-7 (1.3 µg/ml), 1-2C3 (2.5 µg/ml), and 26-2H11 (10 µg/ml) for 1 h at 37°C. Normal mouse IgG1 (BD PharMingen) at 10 µg/ml was used as a negative control. Binding of Ab was detected using 3-amino-9-ethylcarbazole (Histostain-SP; Zymed Laboratories), which yields a red signal in positive cells. All immunostained tissue sections and cells were counterstained with Mayer’s hematoxylin (Sigma-Aldrich).

Assessment of biological activity of rsG1 and rsG2

PBMC were cultured in the absence or presence of human rIFN- (100 U/ml; Genzyme Diagnostics, Cambridge MA) and were simultaneously treated with PBS, rsG1, or rsG2 (50 nM) for 6 h at 37°C for Northern analyses or 12 and 24 h for immunoblots. The cells were then lysed in TRIzol reagent for RNA extraction and tested by Northern blotting or were solubilized in radioimmunoprecipitation buffer (PBS containing 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, PMSF, aprotinin, leupeptin, and EDTA) and tested by immunoblotting. These same lysates were used for apoptosis assays.

Northern blot analysis

PBMC were incubated in serum-free lymphocyte medium (AIM-V; Invitrogen) alone (control), or with 100 U/ml of human rIFN- (Genzyme Diagnostics), with or without rsG1 or rsG2 (50 nM) for 6 h. The RNA was extracted by TRIzol reagent (Invitrogen) and quantified as before. Northern blots were performed, as previously described, with 5 µg of total RNA loaded into each lane of a 1% agarose-formaldehyde gel. Human CD8, CD3, and -actin DNA fragments used as probes were generated by RT-PCR using PBMC total RNA. The specific primers for CD8 were: upper, 5'-CCG CCG CCA GTC CCA CCT TCC TCC TAT-3'; lower, 5'-ATG GTG GGC GCC GGT GTT GGT GGT C-3'. Primers for CD3 were: upper, 5'-GTG GCC GGG ACC CTG AGA TGG-3'; lower, 5'-ATG TGA AGG GCG TCG TAG GTG TCC-3'. Both sets were selected using Primer Select Laser Gene (DNASTAR, Madison, WI). To identify -actin mRNA, previously reported primers were used (28). The amplicons were gel purified, cloned into pGEM-T Easy Vector System I (Promega), and sequenced for authenticity. Each blot was probed first with CD8, stripped and reprobed for CD3, then stripped and reprobed for -actin. The membranes were hybridized according to the procedures for QuikHyb (Stratagene, Cedar Creek, TX). The blots were exposed to Hyperfilm MP (Amersham Pharmacia Biotech), and the OD of each treatment band and -actin band was quantified by scanning densitometry using Epson 1680 (Epson, Long Beach, CA) and Gel Pro Analyzer (Media Cybernetics, Silver Spring, MD). Values are presented as a fold-change of each treatment integrated OD compared with control group integrated OD after normalization with -actin in two independent experiments.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Generation and biochemical characterization of eukaryotic rsG1 and rsG2

To study the structure and function of sG isoforms, we generated rsG1 and rsG2 in HEK293 cells, then compared the two proteins in biochemical assays. The intron-exon structure of the HLA-G gene is shown in Fig. 1a, as are the proteins derived from transcripts encoding the two soluble isoforms. Recombinant sG1 and rsG2 were produced by subcloning PCR-generated amplicons from S14/8 cells, a mouse fibroblast cell line stably transfected with 6.0 kb of the HLA-G gene (29), into a modified pRC/CMV vector (20) (Fig. 1a), then transfecting these into HEK293 cells. In sequence analyses, we noted that, as expected from splicing in the sG2 protein in which the 1 domain is joined directly to 3, the first amino acid of the 3 domain is changed from Asp (on sG1) to an Asn.

View larger version (30K): [in this window] [in a new window]   FIGURE 1. Expression and characterization of rsG1 and rsG2 in HEK293 cells. a, Schematic depiction of the exon-intron organization of the gene encoding HLA-G, the soluble HLA-G-encoding splice variants generated by intron 4-retaining transcripts, and the protein products generated by the constructs. Upper panel, The arrows indicate the positions of the second pair of primers used for isolation of the transcripts encoding the soluble isoforms. The protein products are represented below the transcripts. FLAG peptide is located in the N terminus, as is the enterokinase (Ek) site used to release the FLAG tag. The 21 aa from intron 4 are located in the C terminus. b, Identification of rsG1 and rsG2 by immunoblotting with anti-FLAG and 16G1, a mAb directed toward HLA-G intron 4 amino acids. c, Binding of 16G1 to the recombinant proteins in ELISA. No binding was detected when normal mouse IgG was substituted for 16G1. d, Immunoblot demonstrating failure of binding of anti-2m to rsG2 (lower panel) with loading identified by anti-FLAG (upper panel). e, Experiment demonstrating that under both native and denaturing conditions, deglycosylation causes faster migration patterns of the recombinant proteins. f, Evidence for disulfide-based dimers in purified rsG1 and rsG2. M = monomers, D = dimers.

Identification of recombinant proteins by immunoblotting.

Fig. 1b shows that both rsG1 and rsG2 were recognized in immunoblots by anti-FLAG and by the mAb, 16G1, which was generated to a unique amino acid sequence encoded by intron 4 (32) and which is present on both sG isoforms. The 16G1 also bound strongly to rsG1 and rsG2 in ELISA experiments (Fig. 1c). Migration patterns were consistent with glycosylation of both rsG1 and rsG2 H chains (Fig. 1b).

Binding of L chain.

As predicted from the primary structure, rsG1 bound anti-₂m in immunoblots, whereas rsG2, lacking the 2 domain, did not (Fig. 1d). Data obtained from ELISA are summarized in Table I. These assays verified the information obtained by immunoblot analysis on ₂m, shown in Fig. 1d. This was further confirmed with the finding in ELISA that the mAb W6/32, which binds to HLA class I H chains only when the H chains are complexed with ₂m (33), failed to bind rsG2 while demonstrating binding to both rsG1 and pooled HLA class I Ag. Isotype-specific IgG1 and IgG2a controls were negative.

Recombinant sG1 and rsG2 were both glycosylated, as illustrated in Fig. 1e, in which incubation with deglycosylating enzymes increased migration on polyacrylamide gels under both native and denatured conditions, but incubation in the absence of enzyme had no effect.

Oligomerization.

As expected for soluble HLA proteins (34), oligomerization was a consistent feature (Fig. 1f). Under nonreducing conditions, rsG1 demonstrated monomers (41 kDa), dimers (82 kDa), and higher molecular mass oligomers. Under reducing conditions, monomers predominated in preparations of rsG1. Recombinant sG2 exhibited a different pattern. Under nonreducing conditions, monomers were absent, although dimers (68 kDa) and oligomers were evident. Reduction of disulfide bonds yielded rsG2 monomers (34 kDa) and dimers.

Production and characterization of mouse mAb to rsG1 and rsG2

Because earlier experiments in our laboratory indicated that production of sG isoforms is a feature of normal pregnancy (17), we initiated experiments to establish the cellular distribution of sG1 and sG2 at the maternal-fetal interface, which might assist in predicting function(s) during pregnancy.

To achieve this goal, we used rsG1 and rsG2 to stimulate mAb in mice. Following generation and cloning by standard techniques, two high-producing clones, 1-2C3 (IgG1, -chain) generated to rsG1 and 26-2H11 (IgG1, -chain) generated to rsG2, were selected for final characterization studies. The results of ELISA used to test for the specificity of the two mAb are shown in Fig. 2 and summarized in Table I. In Fig. 2, 1-2C3 recognized only rsG1, and 26-2H11 strongly bound rsG2. Neither mAb recognized pooled HLA class I Ag. As a positive control, we used the mAb 16G1 (generated to intron 4-encoded sequences present only in soluble isoforms), which reacted strongly with both rsG1 and rsG2, as expected, but not with pooled HLA class I Ag (Fig. 2). Additional ELISA (data not shown) excluded the possibility that either mAb recognized amino acid sequences in the HLA-G intron 4 region or ₂m (Table I).

View larger version (32K): [in this window] [in a new window]   FIGURE 2. Specificity of mAb generated to rsG1 (1-2C3) and rsG2 (26-2H11). An ELISA showing binding patterns of conditioned medium (CM), negative control IgG1, positive control 16G1, and the two new mAb to sG, rsG1, rsG2, and pooled HLA class I Ag (HLA mix). Two concentrations of culture supernatant medium were tested for each mAb, 1/500 and 1/1000. The results are reported as OD450.

Collectively, therefore, the data indicated that 1-2C3 and 26-2H11 identified epitopes on properly glycosylated and folded rsG1 and rsG2 that were not held in common with the other soluble isoform, other HLA class I proteins, ₂m, or HLA-G intron 4 amino acids, and that the binding epitopes did not survive denaturation.

Flow cytometry experiments to detect membrane-bound HLA-G

The ELISA presented above showed definitively that the two new mAb identified the soluble isoforms. To ascertain whether or not the mAb identified the membrane-bound isoforms as well as the soluble isoforms, HEK293 cells were transiently transfected with constructs encoding vector alone, mG1 (obtained from term villous cytotrophoblast cells), and mG2 (obtained from term chorion membrane extravillous cytotrophoblast cells), then tested by flow cytometry.

Table II shows that essentially 100% of the cells in all three transfected cell lines bound W6/32, which was expected because the HEK293 cells express HLA class I Ag. Neither cells transfected with vector alone (negative control) nor cells transfected with mG1 bound 1-2C3 or 26-2H11. Taken together with the ELISA results in which 1-2C3 readily identified rsG1, these results showed that 1-2C3 exclusively identifies the soluble isoform. To verify this interpretation, additional flow cytometry experiments were done with this mAb and 721.221 cells transfected with vector or mG1 (721.221-G1) (data not shown). The 87G, which identifies both mG1 and sG1, was strongly positive with the mG1-transfected cells, but not with vector-transfected cells. The 1-2C3 was negative with both cell lines. Thus, 1-2C3 is subsequently referred to as anti-sG1.

Structural features of the placenta

In early gestation, distinct subpopulations of trophoblast cells are readily identified in tissue sections. These include villous cytotrophoblast cells, syncytiotrophoblast, and extravillous cytotrophoblast cells. Fig. 3a shows a toluidine blue-stained section of a first trimester placental villous with arrows pointing out the villous cytotrophoblast cells and syncytiotrophoblast. The villous cytotrophoblast cells are precursor cells for two differentiation pathways. In one, the villous cytotrophoblast cells lying directly beneath the syncytium merge with the syncytium as the placenta grows. In the second, the villous cytotrophoblast cells proliferate, erupt from the villi, and form columns, which ultimately contact and invade the maternal decidua. A first trimester column in which extravillous trophoblast cells are immunostained with anti-cytokeratin-7 is shown in Fig. 3d. In late gestation, the villous cytotrophoblast cell supply is nearly exhausted and only isolated cells lying directly beneath the syncytium are present in the villous placenta (Fig. 3, b and e). The multinucleated syncytiotrophoblast layer is prominent. The extravillous cytotrophoblast cells have ceased migrating and have formed the defined chorion membrane, which lies between the amnion epithelial membrane encasing the fetus and amniotic fluid, and the modified maternal endometrium termed the decidua (Fig. 3, c and f).

View larger version (75K): [in this window] [in a new window]   FIGURE 3. Features of first trimester placental villi, term placental villi, and the amniochorion membrane. a–c, Show toluidine blue-stained plastic sections of placentas and extraplacental membranes; d–f, illustrate immunostaining with anti-cytokeratin-7 to highlight the trophoblast cells. Closed arrows mark syncytiotrophoblast; open arrows mark cytotrophoblast cells. *, Extravillous cytotrophoblast cells in maternal blood. A, amnion membrane; C, chorion membrane; D, decidua. Original magnifications: a, b, d, and e, x200; c and f, x100.

The next set of experiments was designed to learn whether HLA-G2 was expressed in cells at the maternal-fetal interface, where it could influence immune cell function(s), and, if so, in which specific cells the protein is found. To achieve this goal, immunohistochemical experiments were done using the new mAbs. In each experiment, a matching tissue section was taken onto the same slide and tested with IgG1 as a substitute for the primary mAb. These controls were negative in all experiments (insets to Fig. 4, a–d).

View larger version (72K): [in this window] [in a new window]   FIGURE 4. Immunostaining of placentas and extraplacental membranes with the mAbs identifying sG1 (1-2C3) and m/sG2 (26-2H11). a and e, First trimester placenta; b and f, 16-wk placenta; c and g, term placenta; d and h, term amniochorion. Normal mouse IgG1 controls are shown in insets. a–d, The 1-2C3 was positive with villous cytotrophoblast cells (small closed arrows), aggregates of protein in maternal blood and fibrin (large closed arrows), and essentially all types of cells in the amniochorion. A, amnion membrane; C, chorion membrane; D, decidua. e–h, 26-2H11 (open arrows) was positive only with extravillous cytotrophoblast cells at the leading edge of the trophoblast column in first and second trimester placentas, and in remnants of the migratory cells found in the chorion membrane. Original magnifications x200.

In contrast to the staining pattern exhibited by anti-sG1, anti-m/sG2 reacted exclusively with extravillous cytotrophoblast cells. In early placentas (Fig. 4e), particularly in cells distal to the villi, reactivity was strong. No staining was observed in villous cytotrophoblast cells, syncytiotrophoblast, or villous stromal cells. The same pattern was observed in wk 16 placenta (Fig. 4f). Anti-m/sG2 did not bind to any cells in term villous placentas (Fig. 4g), but stained some chorionic cytotrophoblasts in the amniochorion (Fig. 4h). For both new mAb, the same results were obtained on an additional two samples each of first trimester and term placentas and amniochorion membranes.

The new mAb did not detect Ag in paraformaldehyde-fixed, paraffin-embedded tissues.

Transcription and translation are linked in sG1, but not in m/sG2

Having observed in placentas and extraplacental membranes that in situ sG1 was abundant in many trophoblast cell subpopulations, but that m/sG2 proteins were found only in cells that had emerged from villi, we sought to verify this by testing isolated cells obtained from within the villi (term placental cytotrophoblast cells) and those that had migrated from the villi (term chorion membrane cytotrophoblast cells).

Identification of mRNAs encoding sG proteins.

We tested for sG1 and sG2 mRNAs using RT-PCR. Total cell mRNA was amplified from the isolated cells using primers G45–65 (signal peptide sequence) and G1225 (3' untranslated region), after which the transcripts that were obtained were reamplified using nested primers (intron 4-specific primer and G1225). These experiments identified a single band of 450 bp in purified villous and chorionic extravillous cytotrophoblast cells, as expected (data not shown). Controls for the experiment consisted of testing RNA from the trophoblast-derived choriocarcinoma cell line, Jar (negative control), and the trophoblast-derived choriocarcinoma cell line, JEG-3 (positive control). The results were as expected. The 450-bp amplicon was shown by sequence analysis to comprise sG mRNA.

To investigate which isoform(s) was present, a second PCR from the same sample was performed using A-Blunt/B-XbaI primers. Two amplicons were obtained from villous and extravillous cytotrophoblast cells as well as from JEG-3 cells. The two amplicons of 900 and 700 bp were gel extracted and purified (Fig. 5). Sequence analysis revealed that the 900-bp amplicon was specific for sG1 (containing intact 1, 2, 3, and intron 4 coding regions), and the 700-bp amplicon was specific for sG2 transcripts (1 coding region joined directly to 3 plus coding portion from intron 4).

View larger version (25K): [in this window] [in a new window]   FIGURE 5. Identification of mRNA encoding sG1 and sG2 in purified villous cytotrophoblast cells from term placentas, chorionic extravillous cytotrophoblasts from extraplacental membranes, and JEG-3 choriocarcinoma cells (positive control). Arrows point to 900- and 700-bp amplicons obtained using primers A-Blunt (1 domain) and B-XbaI (intron 4), as presented in the text and in greater detail in Materials and Methods. Sequencing of these transcripts, which were identified in all preparations, confirmed that the 900-bp amplicon corresponded to sG1 and the 700-bp amplicon corresponded to sG2.

The experiments reported above therefore predicted that both villous and extravillous cytotrophoblast cells would contain sG1 and sG2. To test the prediction, purified villous and extravillous cytotrophoblast cells were cytospun onto glass slides and evaluated by immunocytochemistry using anti-sG1 and anti-m/sG2. As shown in Fig. 6a, the morphologically homogenous, purified villous cytotrophoblast cells were >95% positive with cytokeratin-7. As expected from the immunohistochemical stains, but not from the RT-PCR results, these cells bound anti-sG1 (Fig. 6b), but not anti-m/sG2 (Fig. 6c). The purified extravillous cytotrophoblast cells from chorion membranes, which were morphologically diverse as expected (35), comprised >95% trophoblast cells, as determined by testing for cytokeratin-7 (Fig. 6d). The cells were positive with anti-sG1 (Fig. 6e). Furthermore, 15% of these cells exhibited reactivity with anti-G2 (Fig. 6f), as predicted by the immunohistochemical stains on chorion membranes in situ as well as by the RT-PCR. Staining appeared to be both cytoplasmic and membrane (Fig. 6f). Isotype-specific and normal mouse Ig controls for these experiments were negative.

View larger version (30K): [in this window] [in a new window]   FIGURE 6. Identification of HLA-G proteins in preparations of purified cytotrophoblast cells from term placentas and extravillous cytotrophoblast from extraplacental membranes. Purified villous cytotrophoblast cells are shown in a–c; purified extravillous cytotrophoblast cells from the chorion membrane (extravillous cytotrophoblast cells) are shown in d–f. a and d, Essentially all villous and extravillous cytotrophoblast cells were positive with anti-cytokeratin-7, a marker for trophoblast cells. b and e, Both villous and extravillous cells bound anti-sG1 (1-2C3). c and f, Villous cytotrophoblast cells were negative with anti-m/sG2 (26-2H11), whereas 15% of the extravillous cytotrophoblast cells were positive. In all immunostains, isotype-specific control Ab were negative (data not shown). Positive cells showing granular localization of 1-2C3 are pointed out with closed arrows; membrane staining with 26-2H11 is pointed out with open arrows. Original magnifications, x200.

IFN--mediated up-regulation of CD8 mRNA and protein is abrogated in blood mononuclear cells by rsG1 and rsG2

The experiments reported above showed that sG1 is present in many trophoblast subpopulations as well as in blood, and that m/sG2 proteins are prominently expressed in invading trophoblast cells, which ultimately contact the maternal immune cell-containing decidua. To evaluate the postulate that either isoform might affect the ability of maternal CTL to attack the placenta, we investigated the ability of the rsG1 and rsG2 proteins to influence CD8 expression.

Northern blot analyses were done, the results were quantified by scanning densitometer, then CD8 and CD3 signals were normalized to -actin, as shown in Fig. 7a. In two separate experiments, steady state levels of CD8 mRNA were elevated 2-fold when PBMC were incubated with IFN- (100 U/ml). This increase was completely abrogated by cotreatment with either rsG1 or rsG2. By contrast, levels of CD3 mRNA in stripped and rehybridized blots remained unchanged (data not shown).

View larger version (22K): [in this window] [in a new window]   FIGURE 7. IFN- up-regulation of CD8 is abrogated by cotreatment with rsG1 and rsG2. a, Shows the results of two Northern blot experiments in which PBMC were cultured for 6 h in the absence (control) or presence of IFN-, with or without 50 nM of rsG1 or rsG2. Data are shown as fold increase following analysis by scanning densitometer. b, Shows an immunoblot to detect changes in relative protein levels of expression at 12 and 24 h posttreatment. Note that at both 12 and 24 h posttreatment with IFN-, CD8 is up-regulated and that this up-regulation is abrogated by the simultaneous addition of either rsG1 or rsG2. CD3 remained unchanged at both time points. c, An apoptosis assay (PARP cleavage) demonstrates that IFN- with or without the addition of rsG1 or rsG2 did not induce cleavage. HeLa cells treated with ectoposide as a positive control demonstrated PARP cleavage.

rsG1 and rsG2 do not induce apoptosis in PBMC

Because of a previous study in which partially purified, soluble HLA-G1 from tumor cells reportedly killed PHA-activated T cells (14), we investigated the possibility that the reduction we observed in CD8 mRNA and protein might result from cell death via the apoptotic pathway. However, as shown in Fig. 7c, cleavage products of caspase-3 (CPP32, CASP3) were not observed in either rsG1- or rsG2-treated PBMC. By contrast, HeLa cells used as a positive control and treated with etoposide demonstrated apoptosis-associated activation of caspase-3.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
This study is the first to report: 1) the generation of rsG1 and rsG2 in eukaryotic cells, 2) biochemical characteristics of the proteins, 3) differential expression of sG1 and m/sG2 in placentas with exclusive expression of m/sG2 by invasive cytotrophoblast cells, and 4) evidence that both soluble isoforms of HLA-G decrease CD8 expression on lymphocytes without stimulating cell death.

Comparison of the structural features of the two recombinant proteins showed clearly that rsG1 associates with ₂m, whereas rsG2 does not, which has been the subject of debate (7, 36). The sG1 isoform thus appeared to have the expected L chain, H chain configuration that would facilitate normal peptide binding. By contrast, rsG2 gave evidence of existing naturally as a homodimeric glycoprotein that resembles HLA class II more than HLA class I (7); dimers, but not monomers of rsG2 were identified under nonreducing conditions. However, both isoforms demonstrated oligomerization related to disulfide bonding, as predicted from a recent study that did not distinguish among HLA-G isoforms (37). Whether such oligomerization impairs or facilitates activity remains to be determined, and differential binding of peptides remains unexplored. Because both sG1 and sG2 were abundant in the culture supernatant medium of sG1- and sG2-transfected HEK293 cells, our results disagree with reports (30, 31) that the smaller isoforms, including sG2, are not released from producing cells. These disparate results are likely to be due to the nature of the constructs used for transfections. The Bainbridge construct (30), which is not described in detail, is artificial rather than natural and might not contain all of the elements needed for intracellular transport and secretion.

Although mAb generated to rsG1 (1-2C3) and rsG2 (26-2H11) identified their specific soluble isoform in culture supernatants of transfected HEK293 cells, flow cytometry experiments with mG1- and mG2-transfected HEK293 cells demonstrated that only 1-2C3 was entirely specific for the soluble isoform. The mAb to rsG2 identified mG2 as well as sG2. Epitopes recognized by the two mAb did not survive fixation of tissues with paraformaldehyde, and did not yield signals in sensitive immunoblots, which limits their usefulness to some extent. A number of other mAb to HLA-G1 have been generated, and some, including 87G and 4H84, have been characterized in part (32, 38). Recently, Menier et al. (39) reported a group of anti-HLA-G mAb with characteristics similar to 87G and 4H84 as well as one mAb (G/04) that recognized G1, G2, and sG1 (G5). None of these or the previously reported mAb exclusively identifies the soluble isoform of G1. Furthermore, the specificity of some Ab that are extensively used is questionable. For example, mild acid treatment of 4H84 (38) induces cross-reactivity with HLA class Ia H chains (40). Thus, 1-2C3 is entirely novel. Generation of the mAb to m/sG2 is of even greater import, as none specific for either the soluble or membrane isoform of G2 has been reported to date. The two new mAb turned out to be extremely useful in immunohistochemistry experiments, yielding entirely novel information on the expression of sG1 and m/sG2 at the maternal-fetal interface.

Immunohistochemical studies showed for the first time that sG1 is prominent in placentas, extraplacental membranes, and maternal blood. Use of anti-sG (1-2C3) showed clearly that this protein is located in granules within the cytoplasm of the cells, as would be expected of a protein prepared for export. Bainbridge et al. (30) and Fournel et al. (14) have reported that the sG1 protein is exported from transfected cells, and Solier et al. (41) have reported the same for villous cytotrophoblast cells. This observation was confirmed by our finding of high concentrations of sG1 in supernatant culture medium from our sG1-transfected HEK293 cells. By testing isolated villous and extravillous cytotrophoblast cells, we demonstrated definitively that both subpopulations of trophoblast cells transcribe the messages and translate the specific mRNAs into protein identified by anti-sG1. Identification of sG1 mRNA and protein in villous cytotrophoblast cells was not unexpected; Solier et al. (41) purified villous cytotrophoblast cells and identified sG1 mRNA and protein. However, sG1 mRNA and protein have not been previously identified in extravillous cytotrophoblast cells. Taken together, our results support the idea that cytotrophoblast cells are the major source of sG1 in placentas and extraplacental membranes.

The observation that membrane and/or soluble HLA-G2 is associated exclusively with the invasive trophoblast phenotype was entirely unexpected, as there are no data in the scientific literature on the expression of these isoforms. In immunohistochemical experiments, the mAb to m/sG2 bound only to cytotrophoblast cells in the leading edge of trophoblast columns in first trimester placentas and to some chorionic cytotrophoblast cells, which are derived from the invasive trophoblast cells. This latter finding was confirmed by identifying m/sG2 protein in isolated chorion membrane cells by using 26-2H11. We established that m/sG2 protein is very likely to be produced in the cells, not synthesized in other cells, and internalized by the membrane cells by showing that the cells contain mRNAs encoding both mG2 and sG2. Whether some of the smaller isoforms of HLA-G are able to reach the cell surface or be secreted has been a matter of debate (30, 42). Our results are consistent with those of Riteau et al. (42), who reported membrane insertion of the small isoforms, but not those of Bainbridge et al. (30), who failed to identify membrane insertion. Other results indicate that appropriate processing does occur; soluble isoforms that appear to be mainly sG2 (or free H chains) circulate in maternal blood throughout pregnancy (17). The studies reported in this work show clearly that invasive trophoblast cells are the logical source of this protein.

This highly unexpected finding may have profound biological implications. It will be important to learn, for example, whether or not m/sG2 is aberrantly expressed in diseases of pregnancy that involve reduced cytotrophoblast cell migration such as pre-eclampsia. It is known from studies on women with a genetic alteration resulting in an inability to produce the two isoforms of HLA-G1 that lack of HLA-G1 does not prevent mothers from concluding successful pregnancies (18). We suggested in that report that G2 may substitute when G1 is missing. The probability that G2 can substitute when G1 is missing is supported by the studies of Lila et al. (43), who demonstrated that successful heart transplantation is related to the presence of circulating sG, but is unrelated to the specific isoform.

Yet, expression of G2 may not be critical only to pregnancy; these isoforms might well have a role in migration/invasion or other cellular functions that are not limited to trophoblast. As with HLA-G, the single transcript derived from the lymphocyte CD45 gene is also differentially spliced into messages encoding multiple (nine) isoforms (13). Expression of cell-specific CD45 isoform(s) is related to lymphocyte differentiation, growth, viability, and specific function. Even more interestingly, isoform switching occurs in CD45. For example, stimulation of thymocytes expressing CD45R epitopes results in their switching to CD45R(0) expression and apoptosis. These parallels suggest that switching into expression of m/sG2 might result from an environmental encounter and might influence multiple cytotrophoblast cell activities.

The third major finding in this study was that both rsG1 and rsG2 are biologically active and are fully capable of influencing expression of a critical Ag in CTL, i.e., CD8. Northern blot analysis showed that in a relatively short time period, 6 h, IFN- enhanced steady state levels of CD8 in PBMC. This has not, to our knowledge, been previously reported. CD8 is a subunit of the coreceptor CD8 molecule, which is expressed on the surface of CTL and has an important role in CTL activation and, subsequently, graft rejection (44, 45). Concurrent treatment with either rsG1 or rsG2 abrogated the IFN--induced increase in steady state levels of CD8 mRNA and protein in IFN--activated PBMC. This protocol had no observable effect on CD3 mRNA or protein, and stimulated no apoptosis in treated cells.

Our findings dispute those of Fournel et al. (14), who reported that partially purified sG stimulated Fas/FasL-mediated apoptosis in PHA-stimulated CTL function, but are in accord with those of Wiendl et al. (46) and Le Friec et al. (47), who also failed to identify death of HLA-G-treated cells. Possibly, different conditions of testing are the reason for the disparate results, or perhaps the preparations used in the earlier report (14) contained apoptosis-inducing contaminants. In this study, we show that sG treatment of IFN--activated PBMC simply decreased levels of the coreceptor molecule, CD8, without any apparent killing. The biological implications of HLA-G-induced cell death and simple reduction of a cell surface molecule are entirely different. Our observations suggest that CTL might well remain in the pregnant uterus, but be undetected with the usual anti-CD8 mAb and functionally impaired because of low CD8 (48, 49).

In summary, in this work, we present the first report of cell-specific expression of HLA-G isoforms in human placentas as well as critical information on the biological functions of the two structurally different isoforms. It will be of the utmost importance to establish how each isoform functions in pregnancy as well as in tumors and other types of transplantation (50, 51, 52).

	Acknowledgments

 
We appreciate the gifts of mAb 16G1 and 87G as well as the untransfected and HLA-G1-transfected 721.221 cells from D. Geraghty. We appreciate receiving modified pRC/CMV from B. Hudson; purified HLA mixtures from Dr. Kao; generation of mAb to rsG1 and rsG2 by the Monoclonal Laboratory, University of Illinois at Urbana-Champaign; and support from the University of Kansas Center for Reproductive Sciences and the Kansas Biomedical Research Infrastructure Network.

	Footnotes

 
¹ This work was supported by grants from the National Institute of Child Health and Human Development to J.S.H. (HD26429, HD35859, HD39878) and A.T.F. (HD36759 and HD42280).

² Current address: Department of Genetics, Cytogenetics Laboratory, Children’s Mercy Hospital, 2401 Gillham Road, Kansas City, MO 64108.

³ Address correspondence and reprint requests to Dr. Joan S. Hunt, Department of Anatomy and Cell Biology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Kansas City, KS 66160-7400. E-mail address: jhunt{at}kumc.edu

⁴ Abbreviations used in this paper: GAS, IFN--activated site; ₂m, ₂-microglobulin; HEK, human embryonic kidney; mG1/mG2, membrane HLA-G1/HLA-G2; PARP, poly(ADP-ribose) polymerase; sG1/sG2, soluble HLA-G1/HLA-G2.

Received for publication March 28, 2003. Accepted for publication September 19, 2003.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Geraghty, D. E., B. H. Koller, H. T. Orr. 1987. A human major histocompatibility complex class I gene that encodes a protein with a shortened cytoplasmic segment. Proc. Natl. Acad. Sci. USA 84:9145.[Abstract/Free Full Text]
2. Hunt, J. S., H. T. Orr. 1992. HLA and maternal-fetal recognition. FASEB J. 6:2344.[Abstract]
3. Le Bouteiller, P., V. Mallet. 1997. HLA-G and pregnancy. Rev. Reprod. 2:7.[Abstract]
4. Le Bouteiller, P., A. Blaschitz. 1999. The functionality of HLA-G is emerging. Immunol. Rev. 167:233.[Medline]
5. Chu, W., J. Gao, W. J. Murphy, J. S. Hunt. 1999. A candidate interferon- activated site (GAS element) in the HLA-G promoter does not bind nuclear proteins. Hum. Immunol. 60:1113.[Medline]
6. Hunt, J. S.. 2002. Immunogenetics: genetic regulation of immunity in pregnancy. B. C. J. M. Fauser, ed. Reproductive Medicine: Molecular, Cellular and Genetic Fundamentals, Ch. 9. 1.-17. Parthenon Press, Carnforth.
7. Ishitani, A., D. E. Geragthy. 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.[Abstract/Free Full Text]
8. Fujii. T., A., A. Ishitani, D. E. Geraghty. 1992. A soluble form of the HLA-G antigen is encoded by a messenger ribonucleic acid containing intron 4. J. Immunol. 153:5516.
9. Ellis, S. A., M. S. Palmer, A. J. McMichael. 1990. Human trophoblast and the choriocarcinoma cell line BeWo express a truncated HLA class I molecule. J. Immunol. 144:731.[Abstract]
10. Kovats, S., E. K. Main, C. Librach, M. Stubblebine, S. J. Fisher, R. DeMars. 1990. A class I antigen, HLA-G, expressed in human trophoblasts. Science 39:220.
11. Yelavarthi, K. K., J. L. Fishback, J. S. Hunt. 1991. Analysis of HLA-G mRNA in human placental and extraplacental membrane cells by in situ hybridization. J. Immunol. 146:2847.[Abstract]
12. Ober, C., C. L. Aldrich. 1997. HLA-G polymorphisms: neutral evolution or novel function?. J. Reprod. Immunol. 36:1.[Medline]
13. Trowbridge, I. S., M. L. Thomas. 1994. CD45: an emerging role as a protein tyrosine phosphatase required for lymphocyte activation and development. Annu. Rev. Immunol. 12:85.[Medline]
14. Fournel, S., M. Aguerre-Girr, X. Huc, F. Lenfant, A. Alam, A. Toubert, A. Bensussan, P. Le Bouteiller. 2002. Cutting edge: soluble HLA-G1 triggers CD95/CD95 ligand-mediated apoptosis in activated CD8⁺ cells by interacting with CD8. J. Immunol. 164:6100.
15. Kapasi, K., S. E. Albert, S. M. Yie, N. Zavazava, C. L. Librach. 2000. HLA-G has a concentration-dependent effect on the generation of an allo-CTL response. Immunology 101:191.[Medline]
16. Petroff, M. G., P. Sedlmayr, D. Azzola, J. S. Hunt. 2002. Decidual macrophages are potentially susceptible to inhibition by class Ia and class Ib HLA molecules. J. Reprod. Immunol. 56:3.[Medline]
17. Hunt, J. S., L. Jadhav, W. Chu, D. E. Geraghty, C. Ober. 2000. Soluble HLA-G circulates in maternal blood during pregnancy. Am. J. Obstet. Gynecol. 183:682.[Medline]
18. Ober, C., C. Aldrich, B. Rosinsky, A. Robertson, M. A. Walker, S. Willadsen, M. S. Verp, D. E. Geraghty, J. S. Hunt. 1998. HLA-G1 protein expression is not essential for fetal survival. Placenta 19:127.[Medline]
19. Chu, W., M. E. Fant, D. E. Geraghty, J. S. Hunt. 1998. Soluble HLA-G in human placentas: synthesis in trophoblasts and interferon--activated macrophages but not placental fibroblasts. Hum. Immunol. 59:435.[Medline]
20. Sado, Y., A. Boutaud, M. Kagawa, I. Naito, Y. Ninomiya, B. G. Hudson. 1998. Induction of anti-GBM nephritis in rats by recombinant 3(IV)NC1 and 4(IV)NC1 of type IV collagen. Kidney Int. 53:664.[Medline]
21. Laemmli, U. K.. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680.[Medline]
22. Langat, D. K., P. J. Morales, A. T. Fazleabas, J. M. Mwenda, J. S. Hunt. 2002. Baboon placentas express soluble and membrane-bound Paan-AG proteins encoded by alternatively spliced transcripts of the class Ib major histocompatibility complex gene, Paan-AG. Immunogenetics 54:164.[Medline]
23. Kao, K. J.. 1987. Plasma and platelet HLA in normal individuals: quantitation by competitive enzyme-linked immunoassay. Blood 70:282.[Abstract/Free Full Text]
24. Kennett, R. H., T. J. McKearn, K. B. Bechtol. 1982. Monoclonal Antibodies: A New Dimension in Biological Analyses Plenum Press, New York.
25. Lacy, M. J., E. W. Voss, Jr. 1986. A modified method to induce immune polyclonal ascites fluid in BALB/c mice using Sp2/0-Ag14 cells. J. Immunol. Methods 87:169.[Medline]
26. Phillips, T. A., J. Ni, J. S. Hunt. 2001. Death-inducing tumor necrosis factor (TNF) superfamily ligands and receptors are transcribed in human placentas, cytotrophoblasts, placental macrophages, and placental cell lines. Placenta 22:663.[Medline]
27. Chen, H. L., Y. Yang, X. L. Hu, K. K. Yelavarthi, J. L. Fishback, J. S. Hunt. 1991. Tumor necrosis factor- mRNA and protein are present in human placental and uterine cells at early and late stages of gestation. Am. J. Pathol. 139:327.[Abstract]
28. Nandan, D., N. E. Reiner. 1997. TGF- attenuates the class II transactivator and reveals an accessory pathway of IFN- action. J. Immunol. 158:1095.[Abstract]
29. Yang, Y., D. E. Geraghty, J. S. Hunt. 1995. Cytokine regulation of HLA-G expression in human trophoblast cell lines. J. Reprod. Immunol. 29:179.[Medline]
30. Bainbridge, D. R. J., S. A. Ellis, I. L. Sargent. 2000. The short forms of HLA-G are unlikely to play a role in pregnancy because they are not expressed at the cell surface. J. Reprod. Immunol. 47:1.[Medline]
31. Mallet, V., J. Proll, C. Solier, M. Aguerre-Girr, M. DeRossi, Y. W. Loke, F. Lenfant, P. Le Bouteiller. 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. 61:21.
32. Lee, N., A. R. Malacko, A. Ishitani, M. C. Chen, J. Bajorath, H. Marquardt, D. E. Geraghty. 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.[Medline]
33. Barnstable, C. J., W. F. Bodmer, G. Brown, G. Galfre, C. Milstein, A. F. Williams, A. Ziegler. 1978. Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens: new tools for genetic analysis. Cell 14:9.[Medline]
34. Zavazava, N.. 1998. Soluble HLA molecules: biological significance and clinical implications. Mol. Med. Today 4:116.[Medline]
35. Benirschke, K., P. Kaufmann. 2000. Anatomy and pathology of the placental membranes. Pathology of the Human Placenta 295. Springer-Verlag, New York.
36. Rouas-Freiss, N., R. E. Marchal, M. Kirszenbaum, J. Dausset, E. D. Carosella. 1997. The 1 domain of HLA-G1 and HLA-G2 inhibits cytotoxicity induced by natural killer cells: is HLA-G the public ligand for natural killer cell inhibitory receptors?. Proc. Natl. Acad. Sci. USA 94:5249.[Abstract/Free Full Text]
37. Boyson, J. E., R. Erskine, M. C. Whitman, M. Chiu, J. M. Lau, L. A. Koopman, M. M. Valter, P. Angelisova, V. Horejsi, J. L. Strominger. 2002. Disulfide bond-mediated dimerization of HLA-G on the cell surface. Proc. Natl. Acad. Sci. USA 99:16180.[Abstract/Free Full Text]
38. McMaster, M. T., C. L. Librach, Y. Zhou, K. H. Lim, M. J. Janatpour, R. DeMars, S. Kovats, C. Damsky, S. J. Fisher. 1995. Human placental HLA-G expression is restricted to differentiated cytotrophoblasts. J. Immunol. 154:3771.[Abstract]
39. Menier, C., B. Saez, V. Horejsi, S. Martinozzi, I. Krawice-Radanne, S. Bruel, C. Le Danff, M. Reboul, I. Hilgert, M. Rabreau, et al 2003. Characterization of monoclonal antibodies recognizing HLA-G or HLA-E: new tools to analyze the expression of nonclassical HLA class I molecules. Hum. Immunol. 64:315.[Medline]
40. Polakova, K., J. R. Bennink, J. W. Yewdell, M. Bystricka, E. Bandzuchova, G. Russ. 2003. Mild acid treatment induces cross-reactivity of 4H84 monoclonal antibody specific to nonclassical HLA-G antigen with classical HLA class I molecules. Hum. Immunol. 64:256.[Medline]
41. Solier, C., M. Aguerre-Girr, F. Lenfant, A. Campan, A. Berrebi, V. Rebmann, H. Grosse-Wilde, P. Le Bouteiller. 2002. Secretion of pro-apoptotic intron 4-retaining soluble HLA-G1 by human villous trophoblast. Eur. J. Immunol. 32:3576.[Medline]
42. Riteau, B., N. Rouas-Freiss, C. Menier, P. Paul, J. Dausset, E. D. Carosella. 2001. HLA-G2, -G3, and -G4 isoforms expressed as nonmature cell surface glycoproteins inhibit NK and antigen-specific CTL cytolysis. J. Immunol. 166:5018.[Abstract/Free Full Text]
43. Lila, N., C. Amrein, R. Guillemain, P. Chevalier, C. Latremouille, J. N. Fabiani, J. Dausset, E. D. Carosella, A. Carpentier. 2002. Human leukocyte antigen-G expression after heart transplantation is associated with a reduced incidence of rejection. Circulation 105:1949.[Abstract/Free Full Text]
44. Mason, D.. 1988. The roles of T cell subpopulations in allograft rejection. Transplant. Proc. 20:239.
45. Yoshimura, R., J. Chargui, A. Aitouche, P. Veyron, J. L. Touraine. 2000. Induction of hyperacute rejection of skin allografts by CD8⁺ lymphocytes. Transplantation 69:1452.[Medline]
46. Wiendl, H., M. Mitsdoerffer, V. Hofmeister, J. Wischhusen, E. H. Weiss, J. Dichgans, H. Lochmuller, R. Hohlfeld, A. Melms, M. Weller. 2003. The non-classical MHC molecule HLA-G protects human muscle cells from immune-mediated lysis: implications for myoblast transplantation and gene therapy. Brain 126:176.[Abstract/Free Full Text]
47. Le Friec, G., B. Laupeze, O. Fardel, Y. Sebti, C. Pangault, V. Guilloux, A. Beauplet, R. Fauchet, L. Amiot. 2003. Soluble HLA-G inhibits human dendritic cell-triggered allogeneic T-cell proliferation without altering dendritic differentiation and maturation processes. Hum. Immunol. 64:752.[Medline]
48. Sewell, A. K., U. C. Gerth, D. A. Price, M. A. Purbhoo, J. M. Boulter, G. F. Gao, J. I. Bell, R. E. Phillips, B. K Jakobsen. 1999. Antagonism of cytotoxic T-lymphocyte activation by soluble CD8. Nat. Med. 5:399.[Medline]
49. Purbhoo, M. A., J. M. Boulter, D. A. Price, A. L. Vuidepot, C. S. Hourigan, P. R. Dunbar, K. Olson, S. J. Dawson, R. E. Phillips, B. K. Jakobsen, et al 2001. The human CD8 coreceptor affects cytotoxic T cell activation and antigen sensitivity primarily by mediating complete phosphorylation of the T cell receptor chain. J. Biol. Chem. 276:32786.[Abstract/Free Full Text]
50. Wiendl, H., M. Mitsdoerffer, V. Hofmeister, J. Wischhusen, A. Bornemann, R. Meyermann, E. H. Weiss, A. Melms, M. Weller. 2002. A functional role of HLA-G expression in human gliomas: an alternative strategy of immune escape. J. Immunol. 168:4772.[Abstract/Free Full Text]
51. Wiendl, H., M. Mitsdoerffer, V. Hofmeister, J. Wischhusen, A. Bornemann, R. Meyermann, E. H. Weiss, A. Melms, M. Weller. 1999. HLA-G expression protects porcine endothelial cells from xenogeneic cytotoxicity mediated by human natural killer cells. Transplant. Proc. 31:953.[Medline]
52. Ibrahim, E. C., N. Guerra, M. J. Lacombe, E. Angevin, S. Chouaib, E. D. Carosella, A. Caignard, P. Paul. 2001. Tumor-specific up-regulation of the nonclassical class I HLA-G antigen expression in renal carcinoma. Cancer Res. 61:6838.[Abstract/Free Full Text]

This article has been cited by other articles:

				E. D. Carosella, B. Favier, N. Rouas-Freiss, P. Moreau, and J. LeMaoult Beyond the increasing complexity of the immunomodulatory HLA-G molecule Blood, May 15, 2008; 111(10): 4862 - 4870. [Abstract] [Full Text] [PDF]

				A. Naji, S. Le Rond, A. Durrbach, I. Krawice-Radanne, C. Creput, M. Daouya, J. Caumartin, J. LeMaoult, E. D. Carosella, and N. Rouas-Freiss CD3+CD4low and CD3+CD8low are induced by HLA-G: novel human peripheral blood suppressor T-cell subsets involved in transplant acceptance Blood, December 1, 2007; 110(12): 3936 - 3948. [Abstract] [Full Text] [PDF]

				B A Quinn, M A Hayes, R O Waelchli, M W Kennedy, and K J Betteridge Changes in major proteins in the embryonic capsule during immobilization (fixation) of the conceptus in the third week of pregnancy in the mare Reproduction, July 1, 2007; 134(1): 161 - 170. [Abstract] [Full Text] [PDF]

				P. Fons, S. Chabot, J. E. Cartwright, F. Lenfant, F. L'Faqihi, J. Giustiniani, J.-P. Herault, G. Gueguen, F. Bono, P. Savi, et al. Soluble HLA-G1 inhibits angiogenesis through an apoptotic pathway and by direct binding to CD160 receptor expressed by endothelial cells Blood, October 15, 2006; 108(8): 2608 - 2615. [Abstract] [Full Text] [PDF]

				C. Ober, C. Billstrand, S. Kuldanek, and Z. Tan The miscarriage-associated HLA-G -725G allele influences transcription rates in JEG-3 cells Hum. Reprod., July 1, 2006; 21(7): 1743 - 1748. [Abstract] [Full Text] [PDF]

				T. V. F. Hviid HLA-G in human reproduction: aspects of genetics, function and pregnancy complications Hum. Reprod. Update, May 1, 2006; 12(3): 209 - 232. [Abstract] [Full Text] [PDF]

				M. Shiroishi, K. Kuroki, T. Ose, L. Rasubala, I. Shiratori, H. Arase, K. Tsumoto, I. Kumagai, D. Kohda, and K. Maenaka Efficient Leukocyte Ig-like Receptor Signaling and Crystal Structure of Disulfide-linked HLA-G Dimer J. Biol. Chem., April 14, 2006; 281(15): 10439 - 10447. [Abstract] [Full Text] [PDF]

				B.F. Barrier, B.S. Kendall, C.E. Ryan, and K.L. Sharpe-Timms HLA-G is expressed by the glandular epithelium of peritoneal endometriosis but not in eutopic endometrium Hum. Reprod., April 1, 2006; 21(4): 864 - 869. [Abstract] [Full Text] [PDF]

				Y. Q. Yao, D. H. Barlow, and I. L. Sargent Differential Expression of Alternatively Spliced Transcripts of HLA-G in Human Preimplantation Embryos and Inner Cell Masses J. Immunol., December 15, 2005; 175(12): 8379 - 8385. [Abstract] [Full Text] [PDF]

				Z. Tan, A. M. Shon, and C. Ober Evidence of balancing selection at the HLA-G promoter region Hum. Mol. Genet., December 1, 2005; 14(23): 3619 - 3628. [Abstract] [Full Text] [PDF]

				A. Blaschitz, H. Juch, A. Volz, H. Hutter, C. Daxboeck, G. Desoye, and G. Dohr The soluble pool of HLA-G produced by human trophoblasts does not include detectable levels of the intron 4-containing HLA-G5 and HLA-G6 isoforms Mol. Hum. Reprod., October 1, 2005; 11(10): 699 - 710. [Abstract] [Full Text] [PDF]

				P. Le Bouteiller Human villous trophoblast and the lack of intron 4-retaining soluble HLA-G secretion: beware of possible methodological biases Mol. Hum. Reprod., October 1, 2005; 11(10): 711 - 713. [Full Text] [PDF]

				J. S. Hunt and D. E. Geraghty Soluble HLA-G isoforms: technical deficiencies lead to misinterpretations Mol. Hum. Reprod., October 1, 2005; 11(10): 715 - 717. [Full Text] [PDF]

				A. Blaschitz, H. Juch, A. Volz, H. Hutter, and G. Dohr Soluble HLA-G, the discussion is going on! Mol. Hum. Reprod., October 1, 2005; 11(10): 723 - 727. [Full Text] [PDF]

				M. Le Discorde, C. Le Danff, P. Moreau, N. Rouas-Freiss, and E. D. Carosella HLA-G*0105N Null Allele Encodes Functional HLA-G Isoforms Biol Reprod, August 1, 2005; 73(2): 280 - 288. [Abstract] [Full Text] [PDF]

				J. S. Hunt, M. G. Petroff, R. H. McIntire, and C. Ober HLA-G and immune tolerance in pregnancy FASEB J, May 1, 2005; 19(7): 681 - 693. [Abstract] [Full Text] [PDF]

				R. H. McIntire, P. J. Morales, M. G. Petroff, M. Colonna, and J. S. Hunt Recombinant HLA-G5 and -G6 drive U937 myelomonocytic cell production of TGF-{beta}1 J. Leukoc. Biol., December 1, 2004; 76(6): 1220 - 1228. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Morales, P. J. Articles by Hunt, J. S. Search for Related Content PubMed PubMed Citation Articles by Morales, P. J. Articles by Hunt, J. S.