Untitled

Original
Vol. 26 / Núm 3/ Julio-Septiembre 2007: 127-134 Quantifying soluble HLA-G in supernatants
of cultured embryos as a marker of implantation
potential in an assisted reproduction program
S. Martí1, J. Ten2, P. Marcos1, M.J. Artacho1, F.J. Galán3, R. Bernabeu2,4, G. Rubio1 1Immunology, Department of Clinical Medicine, Miguel Hernández University, Sant Joan d’Alacant, 2Bernabeu Institute for Fertility and Gynecology, Alicante, 3Centro de Genética Humana, Alicante, 4Chair of Reproductive Medicine, Miguel Hernández University, Elche, Spain. CUANTIFICACIÓN DE HLA-G SOLUBLE EN SOBRENADANTES DEL CULTIVO DE
EMBRIONES COMO UN INDICADOR DE SU CAPACIDAD DE IMPLANTACIÓN
EN UN PROGRAMA DE REPRODUCCIÓN ASISTIDA
HLA-G desempeña un papel tolerogénico en la interfase materno- HLA-G plays a tolerogenic function at the maternal-fetal interface.
fetal. En programas de reproducción asistida, en los que se cultivan embrio- Detection of the soluble isoforms of HLA-G (sHLA-G) in culture medi- nes in vitro, ha cobrado interés la detección de las isoformas solubles de um derived from embryos grown in vitro, although technically complex, esta molécula (sHLA-G) en el medio en el que se han crecido los embrio- has gained interest in assisted reproduction programs because of an appar- nes. Aunque la determinación es compleja, tiene interés por su aparente ent relationship with embryo competence. Here, an amplified ELISA was relación con la idoneidad de dichos embriones. En el presente trabajo, designed to measure sHLA-G at the concentrations expected in embryo se ha puesto a punto un ensayo ELISA amplificado para medir sHLA-G supernatants using a limiting-dilution assay of the choriocarcinoma JEG- a las concentraciones esperables en dichos sobrenadantes. Como modelo 3 cell line as a surrogate model. With this ELISA approach, 111 single comparativo se ha utilizado el cultivo a dilución límite de la línea de corio- embryo culture supernatants, collected 44-48 hours after intracytoplas- carcinoma JEG-3. Con el ELISA desarrollado se han analizado retrospec- mic sperm injection, were retrospectively analysed and levels correlated tivamente 111 sobrenadantes recogidos a las 44-48 horas de efectuada with pregnancy results. The presence of sHLA-G was demonstrated in 22 fecundación mediante inyección intracitoplasmática del espermatozoide, (19.8%) of the embryo cultures. There was no relationship between sHLA- y los datos se han correlacionado con los resultados reproductivos de G levels and grading of embryo morphology. The reproductive outcome dichos embriones. En 22 de los sobrenadantes (19.8%) se ha detectado of the morphologically normal embryos that were transferred to the sHLA-G. No se ha encontrado relación entre los niveles de sHLA-G y el women uterus (2-3 per patient) was as follows: in the group of women in grado morfológico de los embriones. Los resultados reproductivos de los which all transferred embryos were sHLA-G negative, the pregnancy and embriones de morfología normal que fueron transferidos al útero de las implantation rates were 29% (4 pregnancies/14 women) and 14% (5 ges- pacientes (2-3 por caso) fueron los siguientes: en el grupo de mujeres que tational sacs/35 embryos transferred), respectively. In contrast, in the recibió únicamente embriones sHLA-G negativos, las tasas de embara- group in which the embryo transfers included at least one sHLA-G pos- zo e implantación fueron, respectivamente, 29% (4 embarazos/14 muje- itive the pregnancy and implantation rates increased to 60% (3/5) and res) y 14% (5 sacos gestacionales/35 embriones transferidos). Por contra, 29% (4/14) respectively. In conclusion, sHLA-G levels in preimplantation en el grupo que recibió al menos un embrión sHLA-G positivo, las tasas embryo supernatants can be quantified and results suggest positive asso- de embarazo e implantación subieron al 60% (3 / 5) y 29% (4/14) res- ciation with pregnancy likelihood. sHLA-G detection seems to be useful pectivamente. En conclusión, es posible cuantificar niveles de sHLA-G en to complement morphology in selecting good quality embryos for increas- sobrenadantes de embriones y los resultados obtenidos sugieren asocia- ing implantation rates and reducing multiple gestations.
ción con la probabilidad de embarazo. La detección de sHLA-G, por tanto,podría ser un buen complemento de la selección morfológica de embrio- KEY WORDS: sHLA-G / ICSI / Embryo culture / Pregnancy.
nes útil para incrementar la tasa de implantación y reducir la de emba-razos múltiples.
PALABRAS CLAVE: sHLA-G / ICSI / Cultivo de embriones / Embarazo.
QUANTIFYING SOLUBLE HLA-G IN SUPERNATANTS OF CULTURED EMBRYOS AS A MARKER OF IMPLANTATION . INTRODUCTION
success rates that support the measurement of this molecule From an immunological standpoint the embryo can be as one of such a candidate test for embryonic quality(15-19).
considered a semi-allogeneic graft, and even completely However, the results of these studies are being questioned allogeneic in cases of oocyte or embryo donation. Contrary by prestigious researchers in fertility as: 1) it is unclear which to artificial transplants, the conceptus is tolerated by maternal of the splicing-derived isoforms of HLA-G mRNA and defence barriers. Several immunosuppressive mechanisms, proteins are expressed by the preimplanted embryo; 2) for some of them still pending clarification, contribute to prevent this reason, there is no agreement on which are the appropriate rejection. Specially protected is the extravillous cytotrophoblast, antibodies to capture the sHLA-G protein isoforms and, fetus-derived placental cells that migrate into the uterine obviously, there is no suitable standard reference reagent; wall and contact with maternal cells. Extravillous and 3) as a final consequence, some of the reported cytotrophoblast cells, besides lacking Human Leukocyte concentrations of sHLA-G in embryo-surrounding media Antigens (HLA)-A, HLA-B and HLA class II, which are the seem artifactual, as they exceed the total expected protein major allograft rejection molecules, express HLA-G(1, 2). This content of a preimplantation embryo(20, 21).
non-classical class Ib antigen is thought to be a critical Here, a retrospective study was carried out trying to tolerogenic molecule for the development of pregnancy.
clarify the relationship between sHLA-G levels in embryo HLA-G is minimally polymorphic (24 alleles identified supernatants and the probability of pregnancy. A biotin- so far) and can be found in seven isoforms resulting from streptavidin amplified ELISA was designed to detect the alternative splicing of a single gene(3-6). Isoforms HLA-G1, full-length HLA-G1 and HLA-5 isoforms in the supernatant -G2, -G3 and -G4 exist as membrane anchored molecules, of limiting dilution cultures of the JEG-3 choriocarcinoma and although only the full-length -G1 isoform is easily cell line as a surrogate model of single embryo culture. A detected at the cell surface, the others are revealed by protein standard was also prepared from HLA-G-transfected intracellular staining(7-9). The isoforms HLA-G5, -G6 and - cells. With these tools, 111 embryo culture supernatants G7 lack the exon 5, which codes the transmembrane domain, collected at 44-48 hours after fertilization by intracytoplasmic and would be consequently released by the cells. However, sperm injection (ICSI) were analyzed for sHLA-G. The results only -G5 and -G6 are detectable as soluble molecules, were correlated with morphology, implantation and pregnancy meanwhile -G7 has been only detected in extracts of transfected cells(10). The HLA-G1 isoform can also be found as a solublemolecule after shedding or proteolytic cleavage contributingto the soluble pool of HLA-G (sHLA-G)(11). Some researchers MATERIALS AND METHODS
have been able to detect HLA-G on the surface ofpreimplantation embryos surplus from in vitro assisted Patients
reproduction techniques (ART), as well as sHLA-G in the Nineteen couples that underwent ICSI due to surrounding culture media(12,13). This finding, besides indicating moderate/severe male factor were performed in our in vitro the involvement of this molecule from the earliest stages of Fertilization Unit at Instituto Bernabeu Alicante. The pregnancy, led these authors to suggest a potential utility requirements to enter the study were: maternal age <39, less of sHLA-G as indicator of embryo quality.
than 3 previous embryonic transfers of fresh embryos, normal Currently, in ART, selection of embryos for transferring uterine cavity, an endometrial thickness ≥ 9 mm on the day to the female uterus is based on morphology examination of human chorionic gonadotropin (hCG) administration at day 2-3 post-fertilization. This procedure is not always and at least two high quality embryos on the second day reliable and 40% of patients who appear to have good quality after fertilization [4 cells, less than 25% of cytoplasmic embryos do not become pregnant during that in vitro fragments, without blastomere multinucleation and equal fertilization cycle(14, 15). Consequently, the development of or similar (<20% difference) cells].
an objective, quantitative and non invasive test to predictthe implantation potential of an embryo is a main objective.
Stimulation protocol and ICSI procedure
The availability of this test would increase pregnancy rates In the previous cycle, oral contraceptives were given. A reducing the number of embryos transferred and would long protocol was used, including leuprolide acetate agonist decrease the prevalence of multiple gestations, which is the (Gonapeptyl Depot; Ferring, Madrid, Spain) in the previous biggest medical problem in this area. Several groups have midluteal phase. After pituitary desensitization was obtained, recently reported data on the correlation between levels a combined protocol, using human follicle stimulating of sHLA-G in embryo culture supernatants and fertility hormone (FSH) (Gonal F; Serono, London, UK) and human menopausal gonadotropin (hMG) (HMG-Lepori; Farma- washing, biotin conjugated mAb W6/32 (Serotec, Oxford, Lepori, Barcelona, Spain), was given. Ovarian response was UK) at a concentration of 0.2 μg/ml in PBS containing 1% monitored by transvaginal ultrasound and plasmatic BSA was added (50 μl/well) and incubated overnight at 4 oestradiol levels. Ovulation was induced with 250 μg of ºC. Wells were washed and streptavidin-horseradish recombinant hCG (Ovitrelle; Serono, London, UK). Oocytes peroxidase (HRP) (Biosource, Camarillo, CA), at a concentration were aspirated 36 hours after hCG administration by a of 0.1 μg/ml in PBS containing 1% BSA was added (50 transvaginal ultrasound-guided needle aspiration under μl/well) and incubated in a humidified chamber at 37 ºC sedation. Surrounding oocyte cumulus and corona radiate with shaking. Plates were washed three times with 300 cells were removed by a brief exposure to 80 IU/mL of μl/well of PBS containing 0.05 % Tween 20 (Sigma) and hyaluronidase (Hyase; Vitrolife, Göteborg, Sweden) followed once with 300 μl/well of PBS. During washes the wells were by gentle pipetting. ICSI was carried out 4 hours after oocyte soaked for 1 min intervals to reduce background. A solution retrieval on a heated stage (Tokai Hit Thermoplate, Model of 3,3’,5,5’-tetramethylbenzidine (TMB) HRP substrate MATS-U505R30, Japan) at 37ºC, which was mounted on an (Sigma) was added (50 μl/well) and incubated for 30 min inverted microscope (Nikon Eclipse TE200, Japan) equipped in the dark at RT. Reaction was stopped by the addition with Hoffmann modulation optics and a Narishige of 50 μl/well of 2N H2SO4 and the optical density (O.D.) micromanipulation system (Narishige, Japan). Microinjection determined at a wavelength of 450 nm with a 690 nm reference was performed according to Vansteirteghem et al(22). Only filter using a microplate reader (Asys Hitech, Eugendorf, metaphase II oocytes were injected and then incubated Austria). A supernatant from the human B lymphoblastoid individually in 50 μl droplets of G1.3 medium (Vitrolife) cell line C1R transfected with HLA-G serially diluted from covered with sterile equilibrated mineral oil (Ovoil; Vitrolife) 100 or 50 to 0.1 U/ml in culture media was used for the at 37ºC in an atmosphere of 6% CO2.
generation of standard calibration curves. The sHLAconcentrations of the samples were determined by logistic Embryonic culture
four-parameter curve fitting using the public domain software 16-18 hours after microinjection, the oocytes that showed Elisa for Windows(23), downloaded from the Center for two pronuclei and 2 polar bodies were considered fertilized Disease Control web site (http://www.cdc.gov). Mean O.D.
and their progress was monitored microscopically until the values of supernatant from untransfected cells and embryo day of transfer (see below). 44-48 hours after ICSI, the original culture media (background) were subtracted to generate culture media (G1.3) were collected and frozen at –20˚C standard calibration curves and to estimate sHLA-G until tested for sHLA-G levels. Two-three high quality concentration of supernatants, respectively. The minimum embryos were transferred intravaginally at day 2 or day 3 detectable concentration was 0.1 U/ml.
post-ICSI and the surplus embryos were frozen. Poor qualityor developmentally arrested embryos were discarded.
Human cell lines Hep G2 (hepatocellular carcinoma), ELISA for determination of sHLA-G levels
C2BBe1 (colorectal adenocarcinoma), U-937 (promonocytic) Maxisorp flat-bottom 96-well microtiter modules (Nunc, and K-562 (erythroleukemia) were obtained from ATCC Roskilde, Denmark) were coated with 50 μl/well of monoclonal (Manassas, VA). The human choriocarcinoma JEG-3 was antibody (mAb) MEM-G/9 (Exbio, Praha, Czech Republic) provided by Dr. Castrillo (Centro de Biologia Molecular at a concentration of 10 μg/ml in 0.05 M carbonate/bicarbonate Severo Ochoa, Madrid, Spain). The HLA-G transfected buffer pH 9.6, for 1 h at 37 ºC in a humidified chamber C1R [B-cell lymphoblastoid line (B-LCL) lacking surface (covered tray) and then overnight at 4 ºC. After this one HLA A and B antigens] was provided by Dr. P. Aparicio and the following incubation steps, wells were washed four (University of Murcia, Murcia, Spain). The B-LCL SuUn times with 300 μl of phosphate buffered saline (PBS) containing was obtained after Epstein-Barr virus transformation.
0.5% caseine (Sigma, St Luis, MO) and blotted dry by Peripheral blood mononuclear cells (PBMC) were obtained inversion on clean paper towels. After washing, the wells from a healthy donor. All the cells were grown in RPMI were blocked with 100 μl of PBS containing 5% bovine serum 1640 with Glutamax (Gibco, UK) supplemented with 10% albumin (BSA) (Sigma) and after 15 min at RT with shaking heat-inactivated fetal bovine serum (PAA, Pasching, Austria) the wells were aspirated off, blocking step repeated and and penicillin-streptomycin (Biowhittaker, Walkersville, washed. Undiluted supernatant samples and properly MD). Supernatants were harvested after 3-4 days of cell diluted controls were added to each well (50 μl) and incubated culture (or until confluence for JEG-3) and frozen at –20 in a humidified chamber for 2h at 37 ºC with shaking. After ºC. Membrane HLA-G expression was confirmed by flow QUANTIFYING SOLUBLE HLA-G IN SUPERNATANTS OF CULTURED EMBRYOS AS A MARKER OF IMPLANTATION . Figure 1. Soluble HLA-G detection in culture supernatants of HLA-G-
Figure 2. Detection of sHLA-G in culture supernatants of small numbers
expressing and HLA-G negative control cell lines. All supernatants were of JEG-3 cells as a sensitivity test for the ELISA method. Cells of the JEG-3 tested at least in duplicate with a MEM-G/9 (capture mAb) and W6/32 biotin choriocarcinoma cell line were seeded by limiting dilution plating in volumes (detection mAb) ELISA. O.D. values + SD are shown. Positive supernatants of 100 μl/well on day 0. Supernatants were collected 48 h later and quantified show the estimated U/ml of sHLA-G. for sHLA-G. As most of the O.D. signals were either lower than the background(culture medium) or in between the background and that of the 0.1 U/mlstandard concentration (i.e. the detection limit of the assay), wells wereconsidered sHLA-G positive only if their O.D. values exceeded the meanO.D. + 3 SD values of background wells. cytometry using the HLA-G-specific mAb MEM-G/9 anda secondary goat anti mouse-PE Conjugated (Dako, Glostrup,Denmark) as described(24).
of ELISA measurements in the supernatants collected after44-48 h culture (Figure 2) showed that the assay was able to detect sHLA-G in all the wells from 4 cell cultures andin approximately half of the cultures containing 1-2 cells Performance of the sHLA-G ELISA
per well. Although the protein secretion rate of a human To determine the specificity and detection limit of our preimplantation embryo can be rather different from that sHLA-G ELISA detection method, the supernatants of of the tumour JEG-3 cells (just for comparison, the diameter different HLA-G expressing- and HLA-G negative cell lines of a human preimplantation embryo can be more than ten were tested. Results (Figure 1) showed a clear signal for times the diameter of a single JEG-3 cell), these results suggest C1R-HLA-G transfected and JEG-3 cell lines supernatants, that the ELISA procedure is appropriate to detect sHLA-G whereas the HLA-class I negative cell line K-562 and the from a single (48h) preimplantation embryo (average 4 HLA-class I positive-HLA-G negative cell lines U-937, Hep G2, C2BBe1, B-LCL SuUn, as well as PBMC showed an O.D.
similar or even below the negative control (medium). For Detection of sHLA-G in supernatants of preimplantation
quantification, a calibration standard was prepared from a diluted supernatant of C1R-HLA-G and was assigned 100 Supernatants from 111 embryo cultures, collected 44-48 UsG/ml (reference units) to which all the results were h after ICSI, were analysed for sHLA-G, and the results were correlated with morphology. The cut-off value to consider Given the disparity of reported results concerning the a supernatant as sHLA-G positive was 0.1 U/ml. sHLA-G presence of sHLA-G in the culture medium of IVF-ICSI could be detected in 22 supernatants (19.8 % of total), with embryos, we next asked whether our ELISA system was sHLA-G levels ranging from 0.1 to 2.6 U/ml (mean 0.99 ± able to detect sHLA-G secretion from a smaller number of 0.78 U/ml) (Figure 3). Ten out of 58 good quality embryos, cells. The JEG-3 cell line, which synthesises all HLA-G as determined by morphology evaluation, were sHLA-G isoforms(25), was seeded using limiting dilution methods at positive (mean 1.02±0.88 U/ml), and twelve out of 53 a range of 1 to 16 cells per well in 100 μl cultures. The results poor quality, discarded embryos were also found to secrete embryo morphology. Interestingly, some of the higher sHLA- G values were found in two opposite situations: embryos with advanced cleavage status (i.e. initiating cavitation)which appear as surplus frozen embryos in Figure 3, and in development arrested, discarded embryos.
sHLA-G in embryo supernatants and reproductive outcome
The results of sHLA-G quantification of embryos that were transferred to the women uterus were next correlatedwith pregnancy and implantation rates (Table I). In the group of 5 women in which the embryo transfers includedat least one sHLA-G positive, 3 patients became pregnant (pregnancy rate 60%), in contrast, in the group of 14 womenin which all transferred embryos were sHLA-G negative, 4 women became pregnant (pregnancy rate 29%). Considering Figure 3. sHLA-G levels in embryo supernatants. Samples were grouped
the number of gestational sacs per embryo transferred, in into ‘good quality’ and ‘discarded’ embryos according to the morphology and the first group of women, which received an average of 2.8 fate of the embryo they belonged to. Good quality embryos were transferred embryos per patient including at least one sHLA-G positive, (2-3 per woman) and surplus embryos were frozen. Transferred embryos 4 sacs resulted from a total of 14 embryos. This means an were in turn grouped in ‘pregnancy’ or ‘no pregnancy’ depending on thefinal gestational result. Biochemical pregnancies were included in the ‘no implantation rate of 28.6 %. In opposition, in the group pregnancy’ group. All the supernatants below the zero line represent undetectable receiving only sHLA-G negative embryos (mean 2.5 embryos per patient) only 5 sacs were seen from 35 embryos transferred(implantation rate 14.3 %). These results strongly suggest sHLA-G (mean 0.97±0.73 U/ml). This means that there is that presence of sHLA-G in embryo supernatants correlated no relationship between sHLA-G levels and grading of TABLE I. sHLA-G levels in culture supernatants of transferred embryos in pregnant and non-pregnant patients (biochemical preg-
nancies are not considered)
sHLA-G1/G5 in embryo
Day of transfer
supernatants (U/ml)
(day post-ICSI)
Clinical pregnancy
QUANTIFYING SOLUBLE HLA-G IN SUPERNATANTS OF CULTURED EMBRYOS AS A MARKER OF IMPLANTATION . DISCUSSION
the same duration and settings and that use the same ELISA In ART, the selection of good quality embryos for approach, but differ in the calibrator, published either the transferring to the female uterus is crucial in order to increase complete absence of sHLA-G in all the embryo cultures(21) or both implantation rates and the number of singleton levels ranging from 1.05 to 37.0 ng/ml(16, 18). What is even pregnancies. Recent studies have reported that sHLA-G more important, if embryo culture conditions were similar in the supernatants from 2 to 3-day embryos may be the key but a different antibody combination was used for the ELISA, predictive marker to decide which embryos to transfer or and the calibrator was an HLA-G pattern purified from cryopreserve(15, 17, 18, 26). The results we describe here support placental tissues, the mean concentration found rose to 165 the potential utility of quantifying sHLA-G for improving ng/ml(17). In view of these results, some authors have reminded embryo selection for two reasons. First, it was possible to that a proportion of the reported levels of sHLA-G exceed detect sHLA-G in 19.8 % of embryo cultures. Second, our by far the maximum expected total protein content of a data demonstrate a positive correlation of sHLA-G with preimplantation embryo (i.e. 45-50 ng)(20), that would mean that the detection assays are not calibrated to detect real Systematic investigations claim, however, that sHLA- amounts of sHLA-G. In this regard, Sargent et al.(30) have G is technically undetectable in 2 to 6-day embryo cultures proposed that the overestimation in sHLA-G concentrations and consequently has no utility for improving embryo would result from the significant amounts of contaminating selection(20, 21, 27). These contradictory results raise the question proteins present in some of the calibrators currently in use.
of whether technical differences in the detection procedures The sHLA-G concentrations measured in our study are and/or sHLA-G standards may be critical for quantification referred to a standard in units prepared from the C1R HLA- of sHLA-G. In this regard, we and all the groups that have G transfected cell line, using as background culture media detected the presence of sHLA-G in embryo supernatants, or supernatant from untransfected cells. This semi quantitative except one(17), have used the ELISA or Luminex format with strategy rules out the necessity of determining total protein the pair of antibodies validated by the recent workshop for content and contamination concerns for routine measurements.
measurement of sHLA-G in plasma samples(28). This format The standard in units has been subsequently compared with includes the use of MEM-G/9 as capture antibody (specific a plasma sample quantified in ng/ml, which indicates that for native β2-microglobulin associated shed and cleaved 1 Unit is equivalent to 1.05 ng (data no shown). We have HLA-G1 and -G5) with either W6/32 (a pan-HLA class I found levels of sHLA-G ranging from 0.1 to 2.6 U/ml (i.e.
mAb) or an anti-β2-microglobulin as the reporter antibody.
0.1-2.7 ng/ml), comparable to the levels reported by Rebmann With this ELISA format, we demonstrate the technical et al.(19) using Luminex technology (i.e. 0.04-5.6 ng/ml).
possibility of detecting sHLA-G in 48 h supernatants of 1- An association of sHLA-G levels with morphological 16 cell cultures of the choriocarcinoma cell line JEG-3, which quality of the embryos has not been demonstrated so far(16, could resemble the expected levels in the media surrounding 19). Here, similar proportions of sHLA-G positive embryos embryos 44-48 h post fertilization. Here it should be noted and average sHLA-G concentrations were found in normal that supernatants of JEG-3 cells have been reported to be and abnormal embryos. This finding suggests that measurement sHLA-G negative, and weakly positive after 7-fold of sHLA-G complements but does not replace morphological concentration, using ELISA procedures different from the selection of good quality embryos. Of note, we detected some workshop consensus and with supposed detection limits of the higher values for sHLA-G in development-arrested of 0.15 ng/ml(29). This proves the importance of selecting embryos. This, the lack of correlation with morphology, and the right antibody combination for sHLA-G research.
the fact that transcription of embryonic genome does not The percent of sHLA-G positive embryos in our series is take place earlier than 70 h after oocyte fertilization(31, 32), almost equal to the 19.9 % recently reported by Rebmann points towards an oocitary origin of at least part of the HLA- et al.(19) but considerably lower than the 36.2 % observed by G protein released by the embryo at the 4-8 cell stage. In this Noci et al.(16) or the 43 % reported by Desai et al.(18) in spite regard, several studies have reported HLA-G protein expression of using similar ELISA tests. A possible explanation can be by oocytes and an important presence in the follicular found in the use of distinct sHLA-G preparations to establish fluid that contributes to their maturation(12, 33, 34). We are the cut-off of positivity. Because no accessible, accepted tempted to speculate whether different hormonal status standard reagent exists, every group has used a different and/or hormonal stimulation protocols of the fertility patients calibrator purified from the supernatant of diverse transfected could influence HLA-G mRNA and/or protein content of cells. The poor accuracy of the sHLA-G concentrations reported the oocyte and of the preimplantation embryo, contributing is obvious in that groups that perform embryo cultures of to the final reproductive result. In this regard, a recently identified progesterone response element has been involved line; Dr. P. Aparicio (University of Murcia, Murcia, Spain) in upregulation of HLA-G gene expression in the JEG-3 cell for the C1R-HLA-G cell line; and Mr. P. M. Culatto (University of Manchester) for technical assistance.
Regarding the reproductive outcome of the embryos analyzed in the present study, 28.6% of the embryos successfullyimplanted when the transfers included at least one embryo DISCLOSURES
positive for sHLA-G, in contrast, the proportion fell to 14% The authors have no financial conflict of interest.
when all the embryos were sHLA-G negative. A possiblebias induced by female infertility can be ruled out becausethe infertility of the couples was caused by male factors.
These results are in line with the implantation rates observed in retrospective studies that grouped embryos according to Dept. Medicina Clínica, InmunologíaUniversidad Miguel Hernández de Elche, Campus de Sant Joan the same criteria (38% and 19% implantation rates, respectively)(18)). Interestingly, a prospective cohort study in which patients received either all sHLA-G positive or all Phone: +34-96-591-9447. Fax +34-96-591-9450 negative embryos (2-3 embryos per patient in both groups), showed a higher implantation rate (44%) for the ‘positive’group but a comparable rate (14%) for the ‘negative’ group(15).
REFERENCES
Additionally, and as otherwise expected from these results, 1. Ellis SA, Sargent IL, Redman CWG, McMichael AJ. Evidence for our pregnancy rates (i.e. 60% in the group of patients a novel HLA antigen found on human extravillous trophoblast transferred with at least one positive embryo and 29% with and a choriocarcinoma cell-line. Immunology 1986; 59:595-601.
negative embryos) lie in between the pregnancy rates reported 2. Fanchin R, Gallot V, Rouas-Freiss N, Frydman R, Carosella ED.
Implication of HLA-G in human embryo implantation. Hum after ICSI and morphology plus sHLA-G positive or negative classification: 75-64% for sHLA-G positives compared to 3. Ishitani A, Geraghty DE. Alternative splicing of HLA-G transcripts 36-23% for sHLA-G negatives(15, 18).
yields proteins with primary structures resembling both class-I Thus, with all the previous technical considerations, our and class-II antigens. Proc Natl Acad Sci USA 1992; 89:3947-3951.
data supports the idea that whenever a study is able to detect 4. Kirszenbaum M, Moreau P, Gluckman E, Dausset J, Carosella E.
sHLA-G in embryo supernatants, much better success rates An alternatively spliced form of HLA-G messenger-RNA in human are found for sHLA-G positive embryos than for negative trophoblasts and evidence for the presence of HLA-G transcriptin adult lymphocytes. Proc Natl Acad Sci USA 1994; 91:4209-4213.
ones. Importantly, several embryos whose culture supernatants, 5. Le Bouteiller P, Blaschitz A. The functionality of HLA-G is emerging.
in our hands, did not reveal any sHLA-G were able to implant and achieve pregnancies. Various studies except one(16) have 6. The Anthony Nolan Trust (2007) http://www.anthonynolan.
also reported this finding in their embryo series(15, 17, 18). This would suggest that detectable HLA-G expression or secretion 7. Bainbridge DRJ, Ellis SA, Sargent IL. The short forms of HLA-G at the 4-8 cell stage is not a prerequisite to embryo implantation, are unlikely to play a role in pregnancy because they are not but it might be later on. The presence of sHLA-G at this expressed at the cell surface. J Reprod Immunol 2000; 47:1-16.
cleavage stage may be then a useful epiphenomenon indicating 8. Mallet V, Proll J, Solier C, Aguerre-Girr M, DeRossi M, Loke YW, active protein synthesis, assembly, export to membrane and et al. The full length HLA-G1 and no other alternative form ofHLA-G is expressed at the cell surface of transfected cells. Hum secretion, all of them mandatory activities for the conceptus 9. Ulbrecht M, Maier S, Hofmeister V, Falk CS, Brooks AG, McMaster In summary, the results presented here indicate that MT, et al. Truncated HLA-G isoforms are retained in the endoplasmic sHLA-G detection in supernatants of embryos conceived reticulum and insufficiently provide HLA-E ligands. Hum Immunol after ICSI correlates with successful pregnancy and its objective quantification, along with morphological score, 10. Paul P, Cabestre FA, Ibrahim E, Lefebvre S, Khalil-Daher I, Vazeux can be of value when selecting suitable embryos for transfer. G, et al. Identification of HLA-G7 as a new splice variant of theHLA-G mRNA and expression of soluble HLA-G5, -G6, and -G7transcripts in human transfected cells. Hum Immunol 2000; 61:1138-1149.
ACKNOWLEDGEMENTS
11. Park GM, Lee S, Park B, Kim E, Shin J, Cho K, et al. Soluble HLA- The authors thank Dr. J. L. Castrillo (Centro de Biología G generated by proteolytic shedding inhibits NK-mediated cell Molecular Severo Ochoa, Madrid, Spain) for the JEG-3 cell lysis. Biochem Biophys Res Commun 2004; 313:606-611.
QUANTIFYING SOLUBLE HLA-G IN SUPERNATANTS OF CULTURED EMBRYOS AS A MARKER OF IMPLANTATION . 12. Jurisicova A, Casper RF, MacLusky NJ, Mills GB, Librach CL.
24. Rubio G, Garcia-Garcia J, Galvez J, Lorenzo N, Lajarin F, Garcia- HLA-G expression during preimplantation human embryo Penarrubia P. Determination of parameters that characterize development. Proc Natl Acad Sci USA 1996; 93:161-165.
effector-target conjugation of human NK and LAK cells by flow 13. Menicucci A, Noci I, Fuzzi B, Criscuoli L, Scarselli G, Baricordi O, cytometry. J Immunol Methods 1997; 209:137-154.
et al. Non-classic sHLA class I in human oocyte culture medium.
25. Yao YQ, Barlow DH, Sargent IL. Differential expression of alternatively spliced transcripts of HLA-G in human 14. Van Montfoort APA, Dumoulin JCM, Kester ADM, Evers JLH.
preimplantation embryos and inner cell masses. J Immunol 2005; Early cleavage is a valuable addition to existing embryo selection parameters: a study using single embryo transfers. Hum Reprod 26. Fuzzi B, Rizzo R, Criscuoli L, Noci I, Melchiorri L, Scarselli B, et al. HLA-G expression in early embryos is a fundamental prerequisite 15. Sher G, Keskintepe L, Fisch J, Acacio B, Ahlering P, Batzofin J, et for the obtainment of pregnancy. Eur J Immunol 2002; 32:311-315.
al. Soluble human leukocyte antigen G expression in phase I culture 27. van Lierop MJC, Wijnands F, Loke YW, Emmer PM, Lukassen media at 46 hours after fertilization predicts pregnancy and HGM, Braat DDM, et al. Detection of HLA-G by a specific sandwich implantation from day 3 embryo transfer. Fertil Steril 2005; 83:1410- ELISA using monoclonal antibodies G233 and 56B. Molecular 16. Noci I, Fuzzi B, Rizzo R, Melchiorri L, Criscuoli L, Dabizzi S, et 28. Rebmann V, LeMaoult J, Rouas-Freiss N, Carosella ED, Grosse- al. Embryonic soluble HLA-G as a marker of developmental Wilde H. Report of the Wet Workshop for Quantification of soluble potential in embryos. Hum Reprod 2005; 20:138-146.
HLA-G in Essen, 2004. Hum Immunol 2005; 66:853-863.
17. Yie S, Balakier H, Motamedi G, Librach CL. Secretion of human 29. Juch H, Blaschitz A, Daxbock C, Rueckert C, Kofler K, Dohr G.
leukocyte antigen-G by human embryos is associated with a higher A novel sandwich ELISA for alpha 1 domain based detection of in vitro fertilization pregnancy rate. Fertil Steril 2005; 83:30-36.
soluble HLA-G heavy chains. J Immunol Methods 2005; 307:96- 18. Desai N, Filipovits J, Goldfarb J. Secretion of soluble HLA-G by day 3 human embryos associated with higher pregnancy and 30. Sargent I, Swales A, Ledee N, Kozma N, Tabiasco J, Le Bouteiller implantation rates: assay of culture media using a new ELISA kit.
P. sHLA-G production by human IVF embryos: Can it be measured Reprod Biomed Online 2006; 13:272-277.
reliably? J Reprod Immunol 2007; In press.
19. Rebmann V, Switala M, Eue I, Schwahn E, Merzenich M, Grosse- 31. Tesarik J, Kopecny V, Plachot M, Mandelbaum J, Dalage C, Flechon Wilde H. Rapid evaluation of soluble HLA-G levels in supernatants JE. Nucleologenesis in the human-embryo developing invitro - of in vitro fertilized embryos. Hum Immunol 2007; 68:251- ultrastructural and autoradiographic analysis. Dev Biol 1986; 20. Menezo Y, Elder K, Viville S. Soluble HLA-G release by the human 32. Braude P, Bolton V, Moore S. Human-gene expression 1st occurs embryo: an interesting artefact? Reprod Biomed Online 2006; between the 4-cell and 8-cell stages of preimplantation development.
21. Sageshima N, Shobu T, Awai K, Hashimoto H, Yamashita M, 33. Rizzo R, Melchiorri L, Stignani M, Baricordi OR. HLA-G expression Takeda N, et al. Soluble HLA-G is absent from human embryo is a fundamental prerequisite to pregnancy. Hum Immunol 2007; cultures: A reassessment of sHLA-G detection methods. J Reprod 34. Roberts JM, Taylor CT, Melling GC, Kingsland CR, Johnson PM.
22. Vansteirteghem AC, Nagy Z, Joris H, Jiaen L, Staessen C, Smitz Expression of the CD46 antigen, and absence of class-I MHC J, et al. High fertilization and implantation rates after intracytoplasmic antigen, on the human oocyte and preimplantation blastocyst.
sperm injection. Hum Reprod 1993; 8:1061-1066.
23. Plikaytis BD, Carlone GM. Statistical considerations for vaccine 35. Yie SM, Li LH, Li GM, Xiao R, Librach CL. Progesterone enhances immunogenicity trials Part 1: introduction and bioassay design HLA-G gene expression in JEG-3 choriocarcinoma cells and human and analysis. Vaccine 2005; 23:1596-1605.
cytotrophoblasts in vitro. Hum Reprod 2006; 21:46-51.

Source: http://institutobernabeu.eu/upload/ficheros/publicaciones/quantifying-soluble-hla-g.pdf