Coelomic cells show apoptosis via Fas/FasL system: a comparative study between healthy human pregnancies and missed miscarriages

Human Reproduction Vol.23, No.5 pp. 1159–1169, 2008

doi:10.1093/humrep/den031

Advance Access publication on March 3, 2008

Coelomic cells show apoptosis via Fas/FasL system: a comparative study between healthy human pregnancies and missed miscarriages A. Kaponis1,3, A. Skyrlas1, N. Zagorianakou2, I. Georgiou1, V. Passa2, E. Paraskevaidis1 and G. Makrydimas1 1

Department of Obstetrics and Gynecology, Ioannina University School of Medicine, Ioannina, Greece; 2Department of Pathology, Ioannina University School of Medicine, Ioannina, Greece

3

Correspondence address. Poutetsi 2, 45332 Ioannina, Greece. Tel: þ30-26510-99608/6972-233270; Fax: þ30-26510-99224; E-mail: [email protected]

Keywords: coelomic cells; Fas; FasL; missed miscarriage; pregnancy

Introduction Apoptosis is considered an essential physiological process for the normal development of human embryos either by eliminating abnormal embryonic cells or continuing the immune tolerance to the pregnancy (Kawamura et al., 2001; Kaysli et al., 2003). The Fas/Fas ligand (FasL) system, which is a major regulator in the induction of apoptosis, is associated with the immune tolerance at the deciduas-trophoblast interface (Kauma et al., 1999). Apoptosis of activated maternal immune cells occurs in the human decidua mainly through the Fas/FasL system (Jerzak and Bischof, 2002). This provides an efficient mechanism responsible for killing activated lymphocytes and establishes immune privilege in the placenta (Van Parijs and Abbas, 1998). The increase of apoptosis among glandular and stromal cells during the implantation window allows the trophoblastic invasion into the endometrium

and is related to the increased expression of FasL by endometrial cells (Kaysli et al., 2003; Harada et al., 2004). Aberration of programmed cell death during the early stages of pregnancy can result in pregnancy loss or embryonic maldevelopment (Savion et al., 2002). In human pregnancies, an increase in low molecular weight DNA fragments indicating an increase in the number of apoptotic cells was detected in the trophoblast of spontaneous abortion compared with normal embryonic development trophoblast specimens (Kokawa et al., 1998a, b). The expression levels of apoptosis-related genes in fetal chorionic villi from recurrent pregnancy loss patients were higher than normal controls (Choi et al., 2003). The study of apoptosis in human embryos is constrained by limited material. Most of the studies detecting apoptotic molecules during the first trimester of pregnancy were performed in trophoblast tissue in in vitro conditions or in animal

# The Author 2008. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]

1159

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

BACKGROUND: The Fas/Fas ligand (FasL) system represents one of the main apoptotic pathways controlling placental apoptosis throughout gestation. In the current study, we have examined the Fas/FasL protein expression and the apoptotic incidents of coelomic cells, amniotic cells and trophoblastic tissue in first trimester human pregnancies and missed miscarriages (MM). METHODS: Protein expression was determined by immunofluoresence, western blotting analysis, immunohistochemistry and indirectly by RT –PCR, whereas apoptotic cell death was assessed by in situ DNA fragmentation analysis. RESULTS: Coelomic cells express Fas/FasL proteins, can undergo apoptosis and were the only cells in which apoptosis, Fas protein expression and FasL protein expression were accordingly increased along with gestational age (P 5 0.001, P 5 0.008; P 5 0.012, respectively). In contrast, amniotic cells and trophoblast showed a consistency in the expression levels of Fas/FasL proteins in healthy pregnancies. MM were accompanied by increased Fas/FasL protein expression in all examined samples (P < 0.001). The increase of Fas/ FasL protein expression was accompanied by proportional increase of apoptotic incidents among the coelomic cell population (P 5 0.023, P 5 0.009, respectively), whereas amniotic cells and trophoblast appeared to be resistant to Fas-induced apoptosis. The lowest expression of Fas/FasL proteins and the minimum occurrence of apoptotic incidents were detected in the trophoblast. CONCLUSIONS: These data suggest that there is a different regulation and function of the Fas/FasL system in early human pregnancies. Aberration of the Fas-mediated apoptosis may represent one of the execution-step necessary for pregnancy loss in MM cases.

Kaponis et al.

Materials and Methods Patients and sample collection Samples of coelomic (n ¼ 11) and amniotic fluids (n ¼ 9) and placental tissues (n ¼ 10) were collected from 13 women who underwent elective abortion at 6–11 week of gestation for social reasons (Table I). In addition, coelomic (n ¼ 4), amniotic (n ¼ 4) and placental tissue (n ¼ 5) samples were obtained from five patients with unexplained miscarriages at 7–10 week of gestation (Table I). All pregnancies were dated with transvaginal ultrasound (measurement of crown-rump length). In all MM cases, the fetus was visible but there was no heart activity. None of the women examined had a previous miscarriage. In the pregnancies with MM, the median gestation was 9 weeks (range 7–10), but according to the crown-rump length measurement the estimated interval between embryonic death and sampling was 1–14 (median 8) days. Informed consent was obtained from each patient after the purpose and nature of the study had been fully explained. The Ioannina University Human Investigation Committee approved the project and the use of the collected samples. There were no significant differences among women with healthy pregnancies and women with unexplained miscarriages in regard to their age (P . 0.05). The average age of patients with miscarriages and women with healthy pregnancies were 29 + 6 and 26 + 6 years, respectively. Coelocentesis –amniocentesis –placental tissue collection Coelocentesis was performed as previously described by Makrydimas et al. (1997, 2004). Briefly, the external genitalia and the vagina were

1160

Table I. Sampling of coelomic and amniotic fluid and trophoblastic tissue. No

Gestational age (weeks) according to the LMP*

CRL (mm)

Coelomic

Amniotic

Trophoblast

Normal pregnancy 1 6þ1 2 6þ2 3 6þ3 4 7þ6 5 8þ1 6 8þ1 7 8þ2 8 8þ4 9 9þ1 10 9þ1 11 9þ4 12 9þ6 13 10þ3

4.7 4.8 4.8 15.4 17.2 17.3 18.9 20.3 24.5 25.1 27.8 31.6 41.4

þ þ þ þ þ þ þ þ þ 2 þ þ 2

2 2 2 þ þ 2 þ þ þ þ þ þ þ

þ 2 þ 2 þ 2 þ þ þ þ þ þ þ

Missed miscarriages 1 7þ2 2 8þ1 3 9þ1 4 9þ4 5 9þ5

12.6 11.4 15.7 19.7 29.6

þ þ þ þ 2

2 þ þ þ þ

þ þ þ þ þ

Gestational age of women according to the LMP and to the CRL.

carefully cleansed with an antiseptic solution. Subsequently, transvaginal sonography using a 5-MHz ultrasound transducer (Toshiba SSA-220A, Toshiba, Tokyo, Japan) covered with sterile rubber was performed. The fetal crown-rump length and fetal heart rate were measured and the amniotic membrane, coelomic space and yolk sac were identified. A 20-G needle was introduced transvaginally into the coelomic cavity through a guide attached to the transducer and fluid was aspirated. The first sample of 0.2 ml was discarded to avoid contamination with maternal tissue and a new syringe was used to aspirate a further 2–6 ml of fluid. Following coleocentesis, a different 20-G needle was introduced into the amniotic cavity and fluid was aspirated. Again the first 0.2 ml was discarded and the rest was used for the analysis. Coelomic and amniotic fluid samples were collected in pyrogen-free tubes and transferred immediately to the laboratory for centrifugation and further elaboration (see below). Following vacuum-suction curettage, placental tissue specimens were divided in two portions: one fixed in 4% (v/v) paraformaldehyde and subsequently embedded in paraffin, and 4– 5 mm sections were mounted on charges slides for immunohistochemistry techniques and another washed thoroughly with cold 150 mM phosphate-buffered saline (PBS, pH 7.2– 7.4) to remove non-adhering blood cells. Areas rich in chorionic villi were dissected from the membranes and cut into small explants (2–3 mm3) and subsequently transferred to the laboratory for RNA extraction. Total RNA extraction, and reverse transcription Total RNA was extracted from the sample cells using an RNA isolation kit (Promega, Madison, WI, USA) as described by the manufacturer. Total RNA extraction was performed with RNase-free RQ1 DNase (Promega Inc.) at 378C for 30 min. RNA samples were subjected to reverse transcription using Moloney murine leukemia virus (MMLV) reverse transcriptase (Invitrogen, Leek, The Netherlands). Total RNA from HeLa cell line was also isolated with the Promega total RNA isolation kit and used for generating the calibration curve. Reverse transcription took place in a final reaction volume of 20 ml containing 50 units MMLV reverse transcriptase, 2 ml 10 PCR buffer, 5 pmol specific antisense primer (also used in PCR) for

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

models. Interestingly, the incidents of cell death detected among in vitro cultured embryos can be higher than the in vivo ones, while in vitro embryonic development tends to be slower than in vivo (Harlow and Quinn, 1982; Brison and Schultz, 1997). Therefore, the ability to detect apoptotic molecules in in vivo conditions could be of paramount importance. First trimester embryos are surrounded by the coelomic cavity. At Days 11 and 12 post-conception, cells derived from the yolk sac form a fine, loose connective tissue, the extraembryonic mesoderm between the trophoblast externally and the amnion internally. The coelomic cavity is developing during the fourth gestational week within the extraembryonic mesoderm which divides in two layers, the somatic and the splachnic mesoderm. At the sixth gestational week, the coelomic cavity is in direct contact with the villous trophoblast with no anatomical barrier between them, and cells and substances produced in the trophoblast appear in markedly higher concentrations in the coelomic fluid than in the maternal plasma (Jauniaux et al., 2003; Makrydimas et al., 2005). During the first nine weeks of gestation the coelomic cavity is the largest space inside the gestational sac (Jauniaux et al., 2003; Makrydimas et al., 2005). Moreover, cells from the membranes, the yolk sac and the embryo are shed into the coelomic cavity and can be found in the coelomic fluid (Makrydimas et al., 2004). To our knowledge, the current study represents the first experimental attempt to investigate the occurrence of apoptosis in the early embryonic compartments by examining the Fas and FasL expression and the apoptosis incidents in trophoblast cells and cells from amniotic and coelomic fluid. In addition, we performed a comparative analysis of Fas and FasL expression between women with missed miscarriage (MM) and a normal pregnancy (NP).

Coelomic cells are dying via Fas-mediated apoptosis

either Fas, FasL or b-actin, 4 U RNase inhibitor (Roche), 4 mmol/l of each deoxyribonucleotide triphosphate, 6.25 mmol/l MgCl2, and 2 ml RNA template. The Fas, FasL and ß-actin primers used were: FAs, forward 50 –ATGCACACTCTGCGATGAAG–30 and reverse 50 ATG CACACTCTGCGATGAAG–30 (PCR product: 240 bp); FasL forward 50 –GGAATGGGAAGACACATATGGAACTGC–30 and reverse 50 –CATATCTGGCCAGTAGTGCAGTAATTC – 30 (PCR product: 238 bp) and ß-actin forward 50 –AGGCATCCTGAC CCTGAAGTAC – 30 and reverse 50 –TCTTCATGAGGTAGTCTGTCAG –30 (PCR product: 389 bp).

In situ detection of DNA fragmentation DNA fragments in placental tissue sections (5–6 mm) and coelomic and amniotic cell suspensions were detected by the terminal deoxynucleotidyl transferase-mediated deoxy uridine triphosphate nick end labeling technique (TUNEL), using a commercially available apoptosis kit, according to the supplier’s instructions (Apotag kit, Oncor). The method utilizes the free 30 -OH DNA ends generated by

Immunoflourescence Coelomic fluid samples were centrifuged at 768g for 5 min and the pellet was spread on poly-L-Lysine slides (Menzel-Glaser). For the purpose of immunoflourescence, samples corresponding to all the cases were then permeabilized in 0.1% Triton X-100 in PBS for 4 min at room temperature, exposed for 15 min to blocking buffer (10% fetal calf serum in PBS), and subsequently exposed to primary antibodies (Fas c-20, dilution 1:100, FasL N-20, dilution 1:100, Santa Cruz Biotechnologies, CA, USA) for 30 min at room temperature. After two washings with PBS, cells were incubated with fluorescein iso-thiocyanate and/or tetramethyl rhodamin isothiocyanate—conjugated secondary antibodies. The slides were examined in a Leica TCS-SP scanning confocal microscope (Leica Microsystems AG) equipped with an argon/krypton laser and Leica TCS software. Digital recordings were obtained using a Leica TCS NT camera (Leica Microsystems AG). Hoechst-stained nuclei were also detected as a chromatin counterstain for the immunofluorescence. Finally, a combined experiment of immunoflouresence and the TUNEL method was performed in order to identify cells that express one of the apoptotic molecules Fas or FasL while at the same time undergo apoptosis. Immunohistochemistry Immunohistochemical reactivity was tested with antibodies against Fas (anti-Fas [c-20] rabbit polyclonal antibody against the C-terminus of Fas, 1:40 dilution, 2 h incubation at room temperature, Santa Cruz

1161

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Real-time quantitative RT–PCR analysis of Fas and FasL gene expression Analysis of Fas, FasL and of the housekeeping gene b-actin expression was carried out using the LightCycler system (Roche Molecular Biochemicals), following the manufacturer’s instructions. A standard curve for gene expression of Fas, FasL and ß-actin was constructed using total RNA from a HeLa cell line. In brief, 10-fold dilutions of RNA in water were made to give 10 different calibrators, each with a known amount of HeLa RNA. For all products, a standard curve was generated and included in each PCR run. PCR amplification for analyzing gene expression was performed in a 20-ml reaction volume. A reaction mixture (Lightcycler DNA Master Hybridization Probes mixture, Roche Molecular Biochemicals) containing reaction buffer, nucleotides and Taq polymerase was used for the PCR according to the supplier’s instructions. Probe, sense primer, antisense primer, 3.2 ml H2O and 8.0 ml complementary DNA were added to this reaction mixture. The reaction conditions consisted of initial denaturation at 958C for 10 min, followed by 40 cycles of denaturation at 948C for 15 s, annealing at 558C for 10 s, and extension at 728C for 6 s immediately followed by the next amplification cycle. We calculated the relative quantification using the ratio between Fas/b-actin and FasL b-actin mRNA to eliminate the constant differences between genes and gene transcripts that might be generated during the process (e.g. differences due to variations in tissue preparation, template extraction, mRNA integrity and PCR). To ensure that no DNA contamination had occurred, we included 10 samples without reverse transcriptase. PCR product specificity was also verified by electrophoresis through a 2% agarose gel. Quantitative analysis was performed using LightCycler software version 4 (Roche Diagnostics). A melt curve analysis was performed to ensure specific amplification. For each target gene, relative levels of expression was normalized using the actin signal (DCt ¼ jCtactin2CtTargetj). Difference in the expression levels between the two variants was determined according to the formula: 2DDCt, where DDCt ¼ jDCt(MM) 2 DCt(NP)s. The differences from the quantitative PCR of individual RNA preparations were statistically analyzed to give mean and SEM. Experiments and results were repeated and measured in triplicates. The software program Statistical Package for the Social Sciences v11.0.2 (SPSS Inc, Chicago, IL, USA) was used for further statistical analysis of gene expression among experimental groups. In all cases normalized expression data were normally distributed and were compared by two-way analysis of variance (ANOVA) and Spearmans’ correlation test. The significance level (P) was set at 0.05.

DNA fragmentation in apoptotic cells. In brief, tissue sections were deparaffinized in xylene and rehydrated through decreasing concentrations of ethanol, followed by washing in PBS. Sections were incubated with 20 mg/ml proteinase K in PBS, followed by quenching of endogenous peroxidase with 3% H2O2 in PBS. Next, the enzyme terminal deoxynucleotidyl transferase (TdT) was used to incorporate digoxigenin-conjugated 20 -deoxyuridine 50 -triphosphate (dUTP) to the ends of DNA fragments. The apoptosis signal was detected by the addition of an anti-digoxigenin antibody conjugated with peroxidase after chromogen development with 3,30 -diaminobenzidine (DAB). Finally, sections were counterstained with hematoxylin, dehydrated in ethanol, cleared with xylene and mounted with coverslips in permamount medium. For the cell suspension samples amniotic and coelomic fluids were centrifuged at 768g for 5 min and the pellet was spread and fixed on poly-L-Lysine slides (Menzel-Glaser, Germany). Next, TdT was used to incorporate digoxigenin-conjugated dUTP to the ends of DNA fragments. The apoptosis signal was detected by the addition of an anti-digoxigenin conjugated fluorescent antibody. Finally, apoptotic cells were detected under a Leica TCS-SP scanning confocal microscope (Leica Microsystems AG, Wetzlar, Germany) equipped with an argon/krypton laser and Leica TCS software. Digital recordings were obtained using a Leica TCS NT camera (Leica Microsystems AG). Morphologically intact TUNEL-(þ) cells and apoptotic cells in hematoxylin–eosin stained slides and cell suspension samples (defined as cells with condensed, hyperchromatic, ring-like, crescentic or beaded chromatin and often surrounded by a clear halo) were considered as positive and are referred to as apoptotic cells. Areas of obvious necrosis were excluded from counting. The number of apoptotic cells was recorded by using the 40 objective lens and by counting the apoptotic cells in at least 10 randomly selected fields, corresponding to a total of 2000 –3000 cells. The apoptotic index was determined as the number of apoptotic cells expressed as a percentage of the total number of counted cells. Reactive lymph nodes served as positive controls. For negative controls, we performed TUNEL method by the omission of the TdT reaction step.

Kaponis et al.

SDS–PAGE and western blotting Amniotic and coelomic cells were lysed in 150 ml of lysis buffer (20 mM Tris, 150 mM NaCl, 1% Triton X-100, 10% glycerol [Sigma], protease inhibitor cocktail [Roche, Mannheim, Germany], pH 7.4). Lysates were diluted 1:1 in sample buffer (62.5 mM Tris, 2% sodium dodecyl sulphate [SDS], 10% glycerol, 1% ß-mercaptoethanol [Sigma] and 0.003% bromophenol blue [Sigma], pH 6.8), boiled for 10 min, followed by separation on 12% SDS-polyacrylamide gels, and transferred onto nitrocellulose membranes (Protan, Schleicher & Schuell GmBH, Dassel, Germany). The membranes were rinsed in Tris-bufferd saline (TBS)-Tween (TTBS: TBS, 0.1% Tween-20 [Merck]) and blocked with 5% non-fat dry milk in TTBS (blocking buffer) for 1 h, followed by overnight incubation at 48C with a polyclonal antibody against human Fas (anti-Fas [c-20], dilution 1:500, Santa Cruz Biotechnology), a polyclonal antibody against FasL (anti-FasL [N-20], dilution 1:500, Santa Cruz Biotechnology) and a mouse monoclonal antibody against human actin (anti-Actin [NH3] sc-58 679, dilution 1:100, Santa Cruz Biotechnology) in blocking buffer. The membranes were rinsed with TTBS and incubated for 1 h with either HRP-conjugated goat anti-rabbit IgG (Nordic Immunology, Tilburg, The Netherlands) diluted 1:10 000 in blocking buffer or with HRP-conjugated rabbit anti-goat IgG (Nordic Immunology) diluted 1:500. Membranes were again rinsed in TTBS and TBS and subsequently the antibody – protein complexes were visualized using a super-signal chemiluminescent substrate (Pierce, Rockford, IL, USA) and exposure to x-ray film (Fuji, Dusseldorf, Germany). The expression levels of Fas and FasL proteins were normalized and relatively quantified according to the expression of actin using a densitometry software (The Discovery Series: Quantity One 1-D Analysis Software, Biorad, USA).

Results Detection of apoptosis Apoptosis was detected in all three types of samples coelomic cells, amniotic cells and placental tissue. In coelomic samples, the TUNEL method revealed a statistically significant increase (P , 0.001, Spearmans’ correlation test, Table II) in TUNEL-(þ) cells along the gestation period (sixth to ninth week) of healthy pregnancies. Specifically, an increase from 1162

Table II. Spearman’s correlation test of TUNEL positive cells, expression of Fas and FasL proteins between different cell types with respect to the weekly progression of pregnancy (week 6– 10). Weeks Coelomic cells

Amniotic cells

Placenta

Weeks TUNEL r ¼ 0.391 P , 0.001 Fas r ¼ 0.652 P ¼ 0.008 FasL r ¼ 0.118 P ¼ 0.012 Weeks TUNEL r ¼ 0.233 P ¼ 0.354 r ¼ 0.238 Fas P ¼ 0.429 FasL r ¼ 0.053 P ¼ 0.640 Weeks TUNEL r ¼ 20.193 P ¼ 0.089 r ¼ 20.266 Fas P ¼ 0.058 r ¼20.115 FasL P ¼ 0.312

TUNEL

Fas

FasL

r ¼ 0.391 P , 0.001

r ¼ 0.652 P ¼ 0.008 r ¼ 0.473 P ¼ 0.023

r ¼ 0.118 P ¼ 0.012 r ¼ 0.541 P ¼ 0.009 r ¼ 0.385 P ¼ 0.022

r ¼ 0.473 P ¼ 0.023 r ¼ 0.541 P ¼ 0.009 r ¼ 0.233 P ¼ 0.354 r ¼ 0.342 P ¼ 0.259 r ¼ 0.211 P ¼ 0.472 r ¼20.193 P ¼ 0.089 r ¼ 0.257 P ¼ 0.085 r ¼ 0.349 P ¼ 0.061

r ¼ 0.385 P ¼ 0.022 r ¼ 0.238 P ¼ 0.429 r ¼ 0.342 P ¼ 0.259 r ¼ 0.485 P ¼ 0.074 r ¼ 20.266 P ¼ 0.058 r ¼ 0.257 P ¼ 0.085

r ¼ 0.053 P ¼ 0.640 r ¼ 0.211 P ¼ 0.472 r ¼ 0.485 P ¼ 0.074 r ¼ 20.115 P ¼ 0.312 r ¼ 0.349 P ¼ 0.061 r ¼ 0.184 P ¼ 0.207

r ¼ 0.184 P ¼ 0.207

5 to 12% of TUNEL-(þ) cells was detected between sixth and ninth week samples (Fig. 1A). Interestingly, considerable differences were detected between the MM samples and the healthy pregnancy samples corresponding to the same gestational age. An increase from 12 to 25% was detected between MM and healthy pregnancy-coelomic samples at 9þ1 weeks of gestation (Fig. 1A and E) (P , 0.001, ANOVA, Table III). Amniotic samples revealed a consistency for responding to the apoptotic stimuli along the gestation period of healthy pregnancies. The modulation of the TUNEL-(þ) cells between sixth and ninth week samples showed no considerable changes. The TUNEL-(þ) percentage of amniotic cells ranged between 0.8 and 1.2% (Fig. 1B). However considerable changes were detected among the MM samples and the healthy pregnancy samples corresponding to the same gestational age. An increase from 1.1 to 8% was detected between MM and healthy pregnancy-amniotic samples at 9þ1 weeks of gestation (P , 0.001, ANOVA, Table III) (Fig. 1B and F). Placental samples also revealed a consistent behavior with respect to the apoptotic incidents. The TUNEL-(þ) cells ranged from 0.4 to 0.7% along the gestation period of healthy pregnancies, and no considerable modulation was detected (Fig. 1C). Once more, considerable changes were detected among the MM samples and the healthy pregnancy samples corresponding to the same gestational age. An increase from 0.5 to 3.5% was detected between MM and healthy pregnancy-placental samples at 9þ1 weeks of gestation (P , 0.001, ANOVA, Table III) (Fig. 1C and G). Finally, a combined analysis of apoptosis considering all three types of samples shows a similar behavior of apoptotic response among the MM incidents (Fig. 1D).

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Biotechnology), Fas-L (anti-FasL [N-20] rabbit polyclonal antibody against the N-terminus of FasL, 1:40 dilution, 2 h incubation at room temperature, Santa Cruz Biotechnology), using the avidin– biotin –peroxidase method, provided by a commercially available kit, according to the manufacturer’s recommendations (Super Sensitive Immunodetection System, BioGenex). Paraffin-embedded placental tissue sections (5–6 mm) were deparaffinized with xylene and rehydrated through a series of descending graded ethanol. Microwave-processing pretreatment for reactivation of the antigenicity was carried out in a 10 mM citrate buffer (pH 6) at 750 W for 15 min, in two cycles. Endogenous peroxidase activity was blocked by incubation for 15 min in 0.3% H2O2 buffer. After the addition of the primary antibody, the biotinylated multi-link secondary antibody and the avidin–biotin –complex with horse-radish peroxidase (HRP) were applied, followed by the addition of the chromogen (DAB and hydrogen peroxide). Finally, slides were counterstained with hematoxylin, dehydrated in ascending ethanol, cleared with xylene, and mounted with coverslips using a permanent mounting medium. Tissue sections of reactive lymph nodes served as positive controls. For negative control we used rabbit non-immune immunoglobulin (Ig) G instead of the primary antibody.

Coelomic cells are dying via Fas-mediated apoptosis

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Figure 1: Histogram presenting the detection of terminal deoxynucleotidyl transferase-mediated deoxy uridine triphosphate nick end labeling technique TUNEL-(þ). (A) coelomic cells between the sixth and ninth week of gestation in normal pregnancy or missed miscarriage (MM), (B) amniotic cells between the sixth and tenth week of gestation and (C) cells of the placental villi between the sixth and tenth week of gestation. (D) Comparative analysis of TUNEL-(þ) cells of all three sample types between the sixth and ninth week of gestation. (E) coelomic cells between healthy pregnancies (upper panel) and MMs at the same gestational period and (F) amniotic cells between healthy pregnancies (upper panel) and MMs at the same gestational period. (G) Detection of TUNEL-(þ) cells (arrows) of the syncytiotrophoblasts and placental villi of: (I, II) normal pregnancies and (III, IV) MMs, respectively (magnification 400) at 8þ1 weeks of gestation

1163

Kaponis et al.

Table III. Analysis of variance between MM and NP of (A) apoptotic incidents (TUNEL þ, %), (B) Fas protein and mRNA expression and (C) FasL protein and mRNA expression (ANOVA). Cell type

W.B./ qRT– PCR

Coelomic Amniotic Placenta Coelomic

W.B. qRT– PCR

Amniotic

W.B qRT– PCR

Placenta

W.B qRT– PCR W.B. qRT-PCR

Amniotic

W.B qRT– PCR

Placenta

W.B qRT– PCR

N

Mean value

SD

SE

P-values

MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP MM NP

5 11 4 9 5 10 5 11 5 11 4 9 4 9 5 10 5 10 5 11 5 11 4 9 4 9 5 10 5 10

17.451 6.582 7.263 0.902 3.352 0.478 5.981 1.612 7.437 2.479 4.523 0.968 4.036 1.152 1.649 1.096 1.786 0.944 4.038 1.495 3.968 2.352 3.440 0.929 3.429 0.721 1.453 0.803 1.530 0.892

12.250 4.769 6.012 0.795 2.841 0.412 5.044 1.398 7.309 2.296 4.012 0.801 3.869 1.002 1.521 0.968 1.599 0.851 3.672 1.241 3.771 2.198 3.263 0.783 3.279 0.556 1.274 0.735 1.406 0.712

1.832 0.785 0.989 0.131 0.476 0.071 0.897 0.277 1.411 0.358 0.685 0.176 0.495 0.183 0.266 0.016 0.287 0.155 0.611 0.209 0.602 0.336 0.989 0.167 0.581 0.096 0.218 0.119 0.284 0.107

,0.001 ,0.001 ,0.001 0.011 0.006 0.035 0.012 0.259 0.431 0.018 0.029 0.024 0.019 0.356 0.291

n, number of cases; MM, missed miscarriages; NP, normal pregnancies; W.B, western blotting; qRT–PCR, quantitative RT–PCR; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxy uridine triphosphate nick end labeling technique.

Fas and FasL expression The expression of Fas and FasL proteins was detected in all three types of samples, coelomic cells, amniotic cells and placental tissue. Immunofluresence and immunohistochemistry revealed a strong membranous/cytoplasmic expression of Fas protein in all samples, mostly in cells that presented apoptotic morphology (Fig. 2A– C). Morphological characteristics of apoptosis such as chromatin condensation, nuclear shrinkage and nuclear integrity were assessed using a Hoechst nuclear counter stain. The combined identification of the TUNEL-(þ) and Fas-expressing coelomic or amniotic cells revealed that 65 and 70% of the apoptotic cellular population, respectively, express the protein (Fig. 2D). Coelomic and amniotic cells showed a strong membranous/ cytoplasmic expression of FasL protein (Fig. 2A – C). However, FasL was almost exclusively detected in cells that were also expressing Fas protein and presented apoptotic morphology. The combined identification of the TUNEL-(þ) and FasL- expressing coelomic or amniotic cells revealed that in both cellular populations all the cells that express the protein were undergoing apoptosis (Fig. 2D). However, only among the coelomic cellular population Fas and FasL levels of expression were significantly positively correlated with the increase of apoptotic incidents (P ¼ 0.023, r ¼ 0.473 and 1164

Modulation of Fas and FasL expression between the sixth and the ninth gestational week of a NP Coelomic cells revealed a small increase of Fas protein levels between the sixth and the ninth week of gestation of a healthy pregnancy. Western blotting analysis showed a statistically significant increase of 0.5-fold per week reaching a 2-fold difference in the expression levels between sixth and ninth week samples (P ¼ 0.008, Spearmans’correlation test, Table II) (Fig. 3A). FasL expression also showed a similar increase along the gestation period indicating a 1.6-fold difference between sixth and ninth week samples (P ¼ 0.012, Spearmans’ correlation test, Table II) (Fig. 3A). Amniotic samples revealed a consistency in the expression levels of both Fas and FasL proteins along the gestation period of healthy pregnancies. The modulation of the expression levels of both Fas and FasL proteins between sixth and ninth week samples showed no considerable changes (Fig. 3B). Placental samples also revealed a consistency in the expression levels of both Fas and FasL proteins along the gestation period of healthy pregnancies (Fig. 3C). The modulation of the expression levels of both Fas and FasL proteins between sixth and ninth week samples showed no considerable changes (Fig. 3C). Modulation of Fas and FasL expression between MM and NP In the coelomic cellular population a high statistically significant increase of Fas and FasL protein levels was detected among the MM samples and the healthy samples corresponding to the same gestational age (at 9þ1 weeks of gestation). Western blotting analysis and RTPCR showed a 4.4- and a 4.6-fold difference, respectively, in protein and mRNA levels of Fas (P ¼ 0.011 and P ¼ 0.006, respectively, ANOVA Table III), and a 2- and 1.7-fold difference, respectively, for FasL (P ¼ 0.018 and P ¼ 0.029, respectively, ANOVA Table III), between MM and healthy pregnancies. In the amniotic cellular population considerable changes of Fas and FasL protein levels were detected among the MM samples and the healthy samples corresponding to the same gestational age. Western blotting analysis and RT – PCR showed a 3.2- and a 3.9-fold increase, respectively, in protein and mRNA levels of Fas (P ¼ 0.035 and P ¼ 0.012, respectively, ANOVA Table III), and a 2.7- and 3-fold increase, respectively, for FasL (P ¼ 0.024 and P ¼ 0.019, respectively,

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Coelomic

MM/NP

P ¼ 0.009 r ¼ 0.541, respectively) and with each other (P ¼ 0.022, r ¼ 0.385) (Table II). Placental tissue showed strong membranous/cytoplasmic Fas expression and moderate membrane/cytoplasmic FasL expression identified in an average of 10 and 5% of the placental vili cellular population, respectively. The expression profile of Fas and FasL proteins that we detected in the placental tissue is in accordance with previous studies that utilize the commonly used Fas (c-20) and FasL (N-20) antibodies which reflect the cytoplasmic as well as the membranous epitope region of the proteins (Uckan et al., 1997; Grobholz et al., 2002) (Fig. 2C).

Coelomic cells are dying via Fas-mediated apoptosis

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Figure 2: (A) Immunofluresence of Fas and Fas ligand (FasL) proteins in coelomic cells, and (B) in amniotic cells; both showing co-expression of the two protein molecules exclusively in cells that presented condensation of chromatin as assessed using Hoechst counter stain. (C) Immunohistochemical detection of Fas and FasL proteins in cells of the placental villi of normal pregnancies and MMs, showing membranous/cytoplasmic expression (magnification 400). (D) Combined identification of the TUNEL-(þ) coelomic cellular population co-expressing Fas and FasL proteins [uniform arrows indicate the proteins while doted arrows indicate the TUNEL-(þ) cells]

1165

Kaponis et al.

ANOVA Table III), between MM and healthy pregnancy amniotic samples at 9þ1 weeks of gestation. In the placental cellular population no considerable changes of Fas and FasL protein levels were detected among the MM samples and the healthy samples corresponding to the same gestational age. Western blotting analysis and RT – PCR showed a 0.7- and a 0.8-fold difference, respectively, in protein and mRNA levels of Fas, and a 0.5- and 0.8-fold difference, respectively, for FasL, between MM and healthy pregnancy placental samples at 9þ1 weeks of gestation. Discussion Fas and FasL expression and apoptosis in coelomic cells To our knowledge this report shows for the first time that coelomic cells express Fas and FasL proteins. Coelomic fluid is essentially an ultrafiltrate of the maternal plasma, the yolk sac and the embryo. Cells from the trophoblast, the membranes, the yolk sac and probably the embryo itself are shed in the coelomic cavity and can be found in the coelomic fluid 1166

(Makrydimas et al., 2004). Immunofluorescense techniques revealed a strong membranous/cytoplasmic expression of Fas protein, whereas FasL protein was also detected in the membrane region of the coelomic cells. The expression of these two apoptotic molecules was higher in the coelomic cells compared with the expression in the amniotic and trophoblast cells. Moreover, the percentage of TUNEL-(þ) cells in the coelomic fluid was higher compared with the amniotic fluid and trophoblast cells. Different studies by Gruslin et al., (2001) and De Falco et al., (2004) have shown that Fas immunosignals were strongest in cytotrophoblasts during the first trimester of pregnancy. An original finding of the current study is that cells in the coelomic fluid showed a more frequent expression of Fas protein compared with the FasL molecule. Since the coelomic cavity is surrounded by cytotrophoblast this finding suggests that coelomic cells expressing Fas protein may originate from the cytotrophoblast layer. We also found that a high percentage of the Fas-(þ) coelomic cells presented with apoptotic morphology, while the in situ DNA fragmentation analysis

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Figure 3: Representative data of densitometry analysis of the western blotting of Fas and FasL proteins of total protein extract from (A) coelomic cells; (B) amniotic cells and (C) placental tissue from healthy pregnancies or MMs

Coelomic cells are dying via Fas-mediated apoptosis

revealed that 60– 70% of the apoptotic coelomic population expressed the Fas protein. Furthermore, our findings indicate that the majority of the coelomic cell population that expressed the FasL protein in the membrane region of the cytoplasm also appeared with morphological characteristics of apoptosis, while the combined identification of the TUNEL-(þ) and FasL expressing coelomic cells revealed that almost all the cells that express the protein were undergoing apoptosis.

Localization of Fas and FasL in trophoblast cells A number of studies have reported the expression of Fas and FasL by villous trophoblast of human early pregnancy (Runic et al., 1996; Hammer and Dor, 2000; Abrahams et al., 2004a, b; Frangsmyr et al., 2004). Although these studies are in agreement for the discordant localization of Fas and FasL in the placenta as well as for the general dissociation of their expression from the incidence of apoptosis, there have been different opinions considering the precise localization of FasL in the villous trophoblast. Some studies have described the FasL expression as cytoplasmic (Hammer and Dor, 2000; Abrahams et al., 2004a, b; Frangsmyr et al., 2004), whereas others have reported membranal expression (Runic et al., 1996). Moreover, there are studies supporting that FasL is expressed by both syncitio- and cytotrophoblast (De Falco et al., 2004; Pongcharoen et al., 2004), whereas others detect FasL in cytotrophoblast only ( Huppertz et al., 1998; Hammer and Dor, 2000). It seems that the expression of Fas/FasL system of trophoblast cells becomes variable in early, NP. In the current study, we used freshly isolated trophoblast cells in order to exclude possible mRNA contamination from other types of placental cells. Our initial RT – PCR approach

Fas and FasL expression and apoptosis in amniotic cells During the first trimester of pregnancy the amniotic cavity encircles the embryo and is surrounded by the coelomic cavity. Cells in the amniotic fluid originate from the fetus, whereas transportation of coelomic cells through the amniotic membrane cannot be excluded. The localization of Fas and FasL protein in the amniotic cells was similar to that noted in the coelomic cells. We observed a strong cytoplasmic expression of Fas and a strong membranal expression of FasL protein. The similar pattern of expression of Fas and FasL in amniotic and coelomic cells support the hypothesis that an active transportation of cells exists between the two cavities. However, western blotting analysis and RT – PCR techniques revealed a consistency in the expression levels of both Fas and FasL proteins and mRNA along the examined gestational period in amniotic samples, whereas in the coelomic samples there was a small increase of Fas and FasL protein levels between the sixth and ninth week of gestation of a healthy pregnancy. Moreover, in contrast with the coelomic cells the amniotic cells showed a consistency regarding the TUNEL-(þ) cells between the sixth and ninth week of gestation corresponding to the consistency of expression of Fas and FasL protein. Trophoblast cells showed a similar pattern to the amniotic cells regarding the Fas and FasL protein expression and the number of apoptotic cells. The latter finding is in agreement with data in studies by Kokawa et al. (1998a, b) confirming that apoptosis in NP does not show any quantitative change from six to eight weeks of gestation. Variation of apoptosis during the first trimester of pregnancy Interestingly among the coelomic cell population we found an increase of apoptotic cells and an increase of the levels of expression of Fas and FasL proteins between the sixth and ninth weeks of gestation in healthy pregnancies. We believe that trophoblast cells shedding in the coelomic cavity are responsible for this increase since they loose the protection of apoptosis-inhibitory molecules (X-linked inhibitor of apoptosis, FLIP) (Aschkenazi et al., 2002; De Falco et al., 2004; 1167

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Fas and FasL expression and apoptosis in trophoblast cells In the placental villi of the same pregnancies a significantly smaller percentage of cells, ranged from 0.4 – 0.7% were apoptotic suggesting that under normal conditions there is a trophoblast resistance to Fas-mediated apoptosis in the first trimester of pregnancy. This finding is in agreement with data from different studies showing that the X-linked inhibitor of apoptosis protects first trimester trophoblast cells from Fas-mediated apoptosis (Aschkenazi et al., 2002; De Falco et al., 2004; Straszewski-Chavez et al., 2004). Moreover, interleukin-10 (IL-10)-induced FLICE (FADD-like IL-1b-converting enzyme)-inhibitory protein (FLIP) expression and activation protects first trimester trophoblast cells from Fas-mediated apoptosis (Aschkenazi et al., 2002). Immunoflourescense techniques in the current study showed that trophoblast cells appeared with a strong cytoplasmic expression of Fas protein and a moderate membrane expression of FasL. We, as others, propose that first trimester trophoblast cells are resistant to Fas-mediated apoptosis, even though they are expressing both Fas and FasL (Aschkenazi et al., 2002; De Falco et al., 2004; Straszewski-Chavez et al., 2004). It is possible that these cells shedding from the trophoblast to the coelomic cavity loose the protective role of these molecules and appear in the coelomic fluid with increased apoptotic index.

confirmed that these cells constitutively express functionally active FasL. A previous study has shown that ex vivo explant and cell culture systems may be prone to execution caspase activation in cytotrophoblast in vitro, whereas freshly obtained tissues do not show activity of these molecules in the cytotrophoblast layer (Huppertz et al., 1999). We did not attempt to discriminate the expression of FasL between the syncytioand cytotrophoblast. In contrast to the findings of other studies (Abrahams et al., 2004a, b; Frangsmyr et al., 2004), we found membranal expression of FasL in 5 – 10% of trophoblast cells in early, NP. A study by Frangsmyr et al. (2004) have supported that the contradictory result regarding the localization of FasL may be due to technical reasons and suggested the use of a monoclonal antibody (G247-4) for the identification of FasL. However, the antibodies used in the current study can detect the cytoplasmic as well as the membranous epitope region of the Fas and FasL proteins (Uckan et al., 1997; Gruslin et al., 2001).

Kaponis et al.

Straszewski-Chavez et al., 2004). Several reports have suggested that apoptosis may initiate degeneration in the syncytiotrophoblast as this layer naturally ages leading to shedding of trophoblastic cells (Huppertz et al., 1998; Mayhew et al., 1999, among others). An in vitro model of human placental trophoblast deportation/shedding has been devised by Abumaree et al. (2006). These authors have provided a direct experimental documentation that shed trophoblast cells are predominantly apoptotic. In addition, since the amniotic cells presented with stable apoptotic index we can hypothesize that following the gestational age a larger amount of trophoblast cells are shedding into the coelomic cavity making it difficult to reach the amniotic cavity at the time of sampling considering also the fact that the amniotic membrane becomes more impermeable along the gestational period. A cytogenetic analysis of the coelomic cell population could be a realistic approach to support this hypothesis. However, due to technical difficulties in culturing the coelomic cells this approach remains impossible.

1168

References Abrahams VM, Kim YM, Straszefski SL, Romero R, Mor G. Macrophage and apoptotic cell clearance during pregnancy. Am J Reprod Immunol 2004a;51:275–282. Abrahams VM, Straszewski-Chavez SL, Guller S, Mor G. First trimester trophoblast cells secrete Fas ligand which induces immune cell apoptosis. Mol Hum Reprod 2004b;10:55– 63. Abumaree MH, Stone PR, Chamley LW. An in vitro model of human placental trophoblast deportation/shedding. Mol Hum Reprod 2006;11:687–694.

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Fas and FasL expression and apoptosis in MM It is well known that the regulation of placental apoptosis is essential for the normal physiology of pregnancy. At the site of implantation there is a constant cell turnover which is necessary for the appropriate growth and function of the placenta (Levy et al., 2002; Abrahams et al., 2004a, b). Choi et al., (2003) have shown that apoptosis-related genes (caspases, Bcl-xL/Bcl-2-associated death promoter [BAD], Bcl-2 interacting domain [BID], Bcl-2-associated X protein, Fas, FasL) were significantly higher in chorionic villi from recurrent pregnancy loss patients than those from healthy pregnancies at both six and eight weeks of gestation. In the current study, we found that the expression of Fas and FasL in MM was higher than in healthy pregnancies at the same gestational age. In addition, the number of TUNEL-(þ) cells was higher in all examined samples in the MM cases compared with healthy pregnancies. In coelomic fluid the increase in the expression levels of Fas and FasL was in accordance with the increase in the number of apoptotic cells. Similarly, in the amniotic fluid we found a consistency between the significantly increased expression of Fas and FasL in MM and the number of cells with apoptotic morphology (8% of amniotic cells were apoptotic). This suggests that the co-expression of Fas and FasL on the cell membrane of amniotic cells triggers the activation of the downstream signaling of the death receptor pathway. On the other hand, the consistency between the response to the apoptotic stimuli and the increase of the Fas/FasL expression we detected holds only a qualitative relation. Our results showed that the increase in the apoptotic index among the MM is almost double compared with the increase of the expression of Fas and FasL proteins. The smallest increase in the expression of Fas and FasL in cases of MM was observed in trophoblast cells. In addition, only 3.5% of trophoblast cells were presenting with apoptotic morphology. The latter finding may suggest that the increase in the number of apoptotic cells in the coelomic fluid (25% of cells were TUNEL-þ) and the amniotic fluid (8% of cells were TUNEL-þ) mainly originated from the fetus and the

yolk sac. This discrepancy suggests that the co-expression of Fas and FasL on the cell membrane does not necessarily ensure the activation of the downstream signaling of the death receptor pathway. Moreover, a soluble form of FasL has been detected in amniotic fluid by Tanaka et al. (1996) and this form can contribute to the increase in the expression levels of FasL in amniotic cells. It is well known that the membrane-bound FasL is converted to the soluble form (sFasL) by the action of certain matrix metalloproteinases. Such secretion of sFasL avoids the cell – cell contact normally associated with Fas-mediated apoptosis and the inflammatory response that appears to occur when membranal FasL is expressed on the cell surface (Chen et al., 1998; Frangsmyr et al., 2004). Another explanation regarding the discordant expression of Fas/FasL and the incidents of apoptosis is that there are two mostly independent pathways leading to apoptosis: one via Fas/FasL signaling, and second via the mitochondrial pathway. Cross-talk between the two pathways is possible under certain conditions, however, they are able to operate independent of each other according to the initiating apoptotic stimulus (Earnshaw et al., 1999; Kumagai et al., 2001). Fas/ FasL signaling under different regulation can actually promote the operation of the mitochondrial pathway via Bid (Kumagai et al., 2001). Caspase-8-mediated cleavage of Bid is translated from the cytosol to the mitochondria where it promotes the exit of cytochrome C that is then associated with apoptosis protease activating factor-1 and the activation of caspase-9, and -3, eventually leading to apoptosis (Hengartner, 2000; Kumagai et al., 2001). In conclusion, in the current study, we evaluated the Fas/ FasL expression and the apoptotic index of coelomic cells, amniotic cells and trophoblast cells during the first trimester of pregnancy in HP and MM. For first time we demonstrated that coelomic cells express Fas and FasL and can undergo apoptosis. We found that it is easy to eliminate a coelomic cell by apoptosis considering that: (i) coelomic cells in normal pregnancies were the only cells where the expression of Fas/FasL and apoptosis increased with gestational age; and (ii) in coelomic cells the increase in the expression of Fas/FasL was associated with an increase in the apoptosis. In contrast, amniotic and trophoblast cells can evade the apoptotic machinery showing a resistance in Fas-mediated apoptosis. The present study provides new insight toward the understanding of apoptosis in the first trimester of pregnancy and sheds light regarding the delicate balance of the Fas/FasL system between cells that contribute to the maintenance of pregnancy.

Coelomic cells are dying via Fas-mediated apoptosis Kayisli UA, Selam B, Guzeloglou-Kayisli O, Demir R, Arici A. Human chorionic gonadotropin contributes to maternal immunotolerance and endometrial apoptosis by regulating Fas-Fas ligand system. J Immunol 2003;171:2305–2313. Kokawa K, Shikone T, Nakano R. Apoptosis in human chorionic villi and deciduas during normal embryonic development and spontaneous abortion in the first trimester. Placenta 1998a;19:21–26. Kokawa K, Shikone T, Nakano R. Apoptosis in human chorionic villi and decidua in normal and ectopic pregnancy. Mol Hum Reprod 1998b;4:87–91. Kumagai K, Otsuki Y, Ito Y, Shibata M-A, Abe H, Ueki M. Apoptosis in the normal human amnion at term, independent of Bcl-2 regulation and onset of labour. Mol Hum Reprod 2001;7:681–689. Levy R, Smith SD, Yusuf K, Huettner PC, Kraus FT, Sadovsky Y, Nelson DM. Trophoblast apoptosis from pregnancies complicated by fetal growth restriction is associated with enhanced p53 expression. Am J Obstet Gynecol 2002;186:1056–1061. Makrydimas G, Georgiou I, Kranas V, Zikopoulos K, Lolis D. Prenatal diagnosis of beta-thalassaemia by coelocentesis. Mol Hum Reprod 1997;3:729–731. Makrydimas G, Georgiou I, Bouba I, Lolis D, Nicolaides KH. Early prenatal diagnosis by coelocentesis. Ultrasound Obstet Gynecol 2004;23:482– 485. Makrydimas G, Vandecruis H, Sotiriadis A, Lakasing L, Spencer K, Nicolaides KH. Coelomic fluid leptin concentration in normal first trimester pregnancies and missed miscarriages. Fetal Diagn Ther 2005;20:406–409. Mayhew TM, Leach L, McGee R, Ismail WW, Myclebust R, Lammiman MJ. Proliferation, differentiation and apoptosis in villous trophoblast at 13-41 weeks of gestation (including observations on annulate lamellae and nuclear pore complexes). Placenta 1999;20:407–422. Pongcharoen S, Searle RF, Bulmer JN. Placental Fas and FasL expression in normal early, term and molar pregnancy. Placenta 2004;25: 321– 330. Runic R, Lockwood CJ, Ma Y, Dipasquale B, Guller S. Expression of FasL by human cytotrophoblast: implications in placentation and fetal survival. J Clin Endocrinol Metab 1996;81:3119–3122. Savion S, Lepsky E, Orenstein H, Carp H, Shepshelovich J, Torchinsky A, Fein A, Toder V. Apoptosis in the uterus of mice with pregnancy loss. Am J Reprod Immunol 2002;47:118– 127. Straszewski-Chavez SL, Abrahams VM, Funai EF, Mor G. X-linked inhibitor of apoptosis (XIAP) confers human trophoblast cell resistance to Fas-mediated apoptosis. Mol Hum Reprod 2004;10:33– 41. Tanaka M, Suda T, Haze K, Nakamura N, Sato K, Kimura F, Motoyoshi K, Mizuki M, Tagawa S, Ohga S et al. Fas ligand in human serum. Nature Med 1996;2:317–322. Uckan D, Steele A, Cherry, Wang BY, Chamizo W, Koutsonikolis A, Gilbert-Barness E, Good RA. 00 Trophoblasts express Fas ligand: a proposed mechanism for immune privilege in placenta and maternal invasion. Mol Hum Reprod 1997;3:655– 662. Van Parijs L, Abbas AK. Homeostasis and self-tolerance in the immune system: turning lymphocytes off. Science 1998;280:243–248. Submitted on July 7, 2007; resubmitted on November 24, 2007; accepted on January 20, 2008

1169

Downloaded from http://humrep.oxfordjournals.org/ by guest on July 1, 2015

Aschkenazi S, Straszewski S, Verwe KM, Foellmer H, Rutherford T, Mor G. Differential regulation and function of the Fas/Fas ligand system in human trophoblast cells. Biol Reprod 2002;66:1853– 1861. Brison DR, Schultz RM. Apoptosis during mouse blastocyst formation: evidence for a role for survival factors including TGF-a. Biol Reprod 1997;56:1088–1096. Chen JJ, Sun Y, Nabel GJ. Regulation of the pro-inflammatory effects of Fas ligand (CD95L). Science 1998;282:1714–1717. Choi HK, Choi BC, Lee SH, Kim JW, Cha KY, Baek KH. Expression of angiogenesis- and apoptosis-related genes in chorionic villi derived from recurrent pregnancy loss patients. Mol Reprod Dev 2003;66:24– 31. De Falco M, Fedele V, Cobellis L, Mastrogiacomo A, Leone S, Giraldi D, De Luca B, Laforgia V, De Luca A. Immunohistochemical distribution of proteins belonging to the receptor-mediated and the mitochondrial apoptotic pathways in human placenta during gestation. Cell Tissue Res 2004;318:599–608. Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem 1999;68:383–424. Frangsmyr L, Baranov V, Nagaeva O, Stendahl U, Kjellberg L, Mincheva-Nillson L. Cytoplasmic microvesicular form of Fas ligand in human early placenta: switching the tissue immune privilege hypothesis from cellular to vesicular level. Mol Hum Reprod 2004;11:35 –41. Grobholz R, Zentgraf H, Kohrmann KU, Bleyl U. Bax, Bcl-2, fas and Fas-L antigen expression in human seminoma: correlation with the apoptotic index. APMIS 2002;110:724–732. Gruslin A, Qiu Q, Tsang BK. X-linked inhibitor of apoptosis protein expression and the regulation of apoptosis during human placental development. Biol Reprod 2001;64:1264–1272. Hammer A, Dohr G. Expression of Fas-ligand in the first trimester and term human placental villi. J Reprod Immunol 2000;46:83 –90. Harada T, Kaponis A, Iwabe T, Taniguchi F, Makrydimas G, Sofikitis N, Paschopoulos M, Paraskevaidis E, Terakawa N. Apoptosis in human endometrium and endometriosis. Hum Reprod Update 2004;10:29– 38. Harlow GM, Quinn P. Development of preimplantation mouse embryos in vivo and in vitro. Aust J Biol Sci 1982;35:187–193. Hengartner HO. The biochemistry of apoptosis. Nature 2000;407:770– 776. Huppertz B, Frank HG, Kingdom JC, Reister F, Kaufmann P. Villous cytotrophoblast regulation of the syncytial apoptotic cascade in the human placenta. Histochem Cell Biol 1998;110:495– 508. Huppertz B, Frank HG, Reister F, Kingdom JC, Korr H, Kaufmann P. Apoptosis cascade progresses during turnover of human trophoblast: analysis of villous cytotrophoblast and syncytial fragments in vitro. Lab Invest 1999;79:1687– 1702. Jauniaux E, Cirigliano V, Adinolfi M. Very early prenatal diagnosis on coelomic cells using quantitative fluorescent polymerase chain reaction. Reprod Biomed Online 2003;6:494–498. Jerzak M, Bischof P. Apoptosis in the first trimester human placenta: the role in maintaining immune privilege at the maternal-fetal interface and in the trophoblast remodeling. Eur J Obstet Gynecol 2002;100:138–142. Kauma SW, Huff TF, Hayes N, Nilkaeo A. Placental Fas ligand expression is a mechanism for maternal immune tolerance to the fetus. J Clin Endocrinol Metab 1999;84:2188–2194. Kawamura K, Fukuda J, Kodama H, Kumagai A, Tanaka T. Expression of Fas and Fas ligand mRNA in rat and human preimplantation embryos. Mol Hum Reprod 2001;7:431– 436.